ENGLISH LANGUAGE SERIES General Editor: Randolph Quirk
Title no:
INVESTIGATING ENGLISH STYLE I
David Srystal and Derek Davy
THE MOVEMENT OF ENGLISH PROSE 2
Ian A. Gordon
A LINGUISTIC GUIDE TO ENGLISH POETRY 4
Geoffrey N. Leech
AN INTRODUCTION TO 7 MODERN ENGLISH WORD-FORMATION
Valerie Adams
COHESION IN ENGLISH 9
M. A. K. Halliday and Ruquaiya Hasan
AN INTRODUCTION TO ENGLISH 10 TRANSFORMATIONAL SYNTAX
Rodney Huddleston
MEANING AND FORM  11
Dwight Bolinger
DESIGNS IN PROSE 12
Walter Nash
STYLE IN FICTION 13
Geoffrey N. Leech and Michael H. Short
THE RHYTHMS OF ENGLISH POETRY 14
Derek Attridge
THE LANGUAGE OF HUMOUR 16
Walter Nash
GOOD ENGLISH AND THE GRAMMARIAN 17
Sidney Greenbaum
RHYTHMIC PHRASING IN ENGLISH VERSE 18
Richard D. Cureton
THE INFINITIVE IN CONTEMPORARY ENGLISH 19
Patrick Duffiey
: I-
Investigating English Style
DAVID CRYSTAL
Professorial Fellow, University College of North Wales, Bangor
DEREK DAVY
Department of English, University of Canterbury, New Zealand
LONGMAN
1
Chapter 4
The Language of Conversation
There are a number of good reasons for choosing relatively informal conversation between educated people as the opening variety for linguistic analysis. Conversation, in the sense described in this chapter, is without doubt the most commonly used kind of English, and consequently a variety which will be more familiar to the vast majority of English-speaking people than any other. We can confidently claim that everyone makes use of this kind of English every day, whereas this claim could not be made of any other variety of English we might want to describe in a book such as this. Also, from the pedagogical viewpoint, the sort of English used in conversational situations, with the extreme kinds of non-fluency mentioned below removed, would seem to be the most useful and least artificial kind to teach foreign students of English as a means of everyday communication. Such practical reasons are quite important.
But we would also point to theoretical and procedural reasons for taking this variety first. It is the least 'marked' kind of situationally-influenced English. By this we mean that, whereas the other varieties in this book are clearly restricted to a particular situation (always to a certain degree specialised) and would be intuitively associated with that situation, conversational English has no comparable situational specificity. It is, situationally speaking, the most neutral kind of English one can find. For this reason, as well as in view of its frequency, it seems the obvious variety to choose for an introductory illustration of our analytic procedure, and as a basic measuring-rod for the language of other situations as and when necessary. We have already discussed the need for comparative statement in stylistic work, and the way in which the existence of a yardstick of some kind may facilitate this (cf note 8,p 91): the present variety would seem to be the most suitable for this purpose, and we shall consequently have cause to refer to it often.1
A relevant procedural reason for choosing this variety first is that conversation is a very convenient kind of English, in that it provides
í.
96
practical analysis
us with a great deal to discuss at all levels of analysis. Very often in stylistics, much of the interest in a text is concentrated at one level -a variety may be primarily distinguished through its phonology, or vocabulary, for example. Here, however, all levels of analysis provide important information about the character of the variety. There seems also to be a much greater flexibility of usage in this variety than in any other: there are fewer restrictions on the kind of structures that may be used, consequently one is liable to find in any extract of conversation that a wider range of contrasts operates at any level than could be expected elsewhere. A further procedural point for a pedagogically orientated book is that as this is the most familiar variety in English, it will be easier for readers to check the linguistic facts presented in our description with their intuitions, and thus make their own assessment of the extent to which our extracts are a representative and helpful sample, than if some less familiar variety had been chosen.
In view of these factors, and the general agreement in linguistics on the primacy of speech in language study, it might seem odd that so little linguistic research has been carried out into this variety of English. There have been occasional informative articles, such as those by Abercrombie and Quirk,2 but these are sporadic in their comments; their main value is to focus attention on certain dominant and yet neglected features of this kind of English. There is little else, and no detailed survey. Far more is known in fact about such varieties of spoken English as advertising or preaching. There is however one very good reason for this lack of information, namely, the procedural difficulty of obtaining reliable data to investigate. It is well-known that most people will behave differently if they are aware of being tape-recorded, and as a result the language they use simply cannot be taken as a reliable sample of spontaneous informal conversation. Even if it seems that they have 'forgotten' about the microphone, the data cannot be trusted. In our experience, there seems to be a cyclic pattern of forgetting and remembering about the microphone, with consequent alterations in the manner of speaking. The only safe way of obtaining data is through the technique of * surreptitious' recording, and this requires a degree of technical preparation which precludes its frequent use. This was the method used for obtaining the material in this chapter: the participants in the conversations, apart from speaker B in extract IV below, were not aware that a recording was being made.3
the language of conversation
97
ŕ
1
The relationship of the speakers to each other in the extracts I, II, and IH which follow is that both are housewives with a general professional background (A in teaching, B in business); they are in the same age-group (mid-thirties) and have known each other for some time. The occasion is that B was invited to A's house for an evening chat over coffee. In extract IV (see p 116), which is a conversation over the telephone, speaker A is female, speaker B male; they are close friends who shared the same university educational background as mature students; A is a housewife, B a lecturer, and both are in their mid-thirties.
The first text consists of three extracts, labelled I, II, and HI, taken from the beginning, middle and end of an evening's conversation respectively. They display some obvious differences - the anecdotal character of II, for example. Consequently, to call the language used throughout the text, along with the extract of telephone conversation, IV, a single variety may be a little premature, though we feel there is sufficient evidence to justify our doing so (seep 116). A clear central area of distinctiveness can be defined, but there are a number of very uncertain marginal issues, which reflect the way in which what is intuitively labelled' conversation' can blend imperceptibly into other varieties that are labelled differently, such as 'discussion', 'talking shop', etc. We shall be looking at this problem again with special reference to extract IV.
'alleg' 'lax' 'dimin' 'alleg piano'
pianiss ' piano'
'piano'
I
A ' you got a ]hCÓld| ' -
B I "nôI ■ 'just a Ibiť tnSNTFFyi cos I'm -
I j"ám cóld| and I'll4 |be all 'right 'once
I've warmed up| - 'do I |look as
'though I've 'got a +cóld| ' A no I |hthought you sôuNDed as 'if you
were
B
A ' I pull your chair up 'close if you wánt| ' |is *it - *(obscured speech}* B *' |yes| ' • |I'll be all 'right in a „MÍNute|* it's [just that I'm ■
10
98       PRACTICAL ANALYSIS
'aUegriss piano wide'
Mow'
'acceľ
'alleg' ' piano' 'acceľ
piano 'piano'
piano narrow
'piano' 'piano' 'piano' * spread'
'low'
piano
'high piano' 'piano'
A (' |i,what have you gót| ')
B tsTupid| I |had s about ttfive 'thousand
b6oks| - 'to |take 'back to 'senate 15
house „YÉsterdayl' - and i got jail
the 'way 'through the „c6LLege| to
I where the car was j'at the I parking
meter at the +ÓTHer end| and [realised
I'd 'left my' • ícóat| in my 20
|„LÓCKer| 'and I *jjust couldn't'*
A
B fáce| 'going I all the way íbáck again| with [this great' ■ 'you know my |"árms were ' aching | 25
A '|nlfl|'-
B 'and I thought |„well| I'll jget
it on tTÚEsdayl - it's a bit
|slLLyj cos I |need it|' * A l\in\ • it's gone |very cóld| 30
[HAsn't it|'
B '|mj---it's)FREEzing|'-
A '|-A|-*rm|(aiyílfl6í«)|'* B *' you're |KNrrTing|* ■ (laughs
quietly for -) ' | what are you 3 5
KNirringj |that's 'not a 'tiny
„GÁRmentj'
A 'jnó|'
B (laughs for -) -
A jno 'it's for mě| but 'it's |very 40 plain)
B 'it's a I lovely *CÓLour|' - -A 'it I is NÍCEJ'
B ' |„yeah| ' - I [never 'di I could |never
■
THE LANGUAGE OF CONVERSATION
take to nKNfrTingl exjcept on these
tdouble o 'needles with t$TRlNG|
'you [kn6w| [that's «my sort of
*'knitting)* A *|„yeah|* B 'it j„GRdws"quickly| A |„yeah| • jl get 'very fed up) B \(jor 4 syllables) the tpR6cess
though| 'do 'you |s£wj • I |used
to sew a ' lot I *' (when) * A *|n6 [i |d6n't)|* B in the |days when I was a 'human
tBEing)'.
'dimin* "narrow"
* affects posh accent' "high forte"
"descend"
'high forte wide' "posh accent" 'piano' "breathy"
'forte'
"alleg"
'high*
'high"
'alleg' "high"
II
B 'and |conver"sAtion"| • |went like +Triis| • [this sort of conversation) am:' - - [1] - ' " ["have you wNdticed " nPRiNcipalj • that am - - the " |boiled eggs I at |sunday *BREAKfast|" '*
A (laughsfor-)
B ' " |always hard| " ' - - ' and [principal SAid| • "|ah [[well]|" ' • the [simple +truth [\\s}\ that • |if you're 'going to 'boil eggs) • |c6MMunally| - they I'hMuST be "hard| • (A and B laugh loudly for —)' and |" every " body twArredl "and she [said you tsEE)" ' — 'you jhave to 'crack the +heAd' of an 'egg I ■ when you [take it 'out of the ipan[ ■ ' I "otherwise" it 'goes
99 45
SO
55
60
65
70
ioo      practical analysis
' alleg'
" high "" narrow * 'allegriss'
•alleg*
'allegriss'
' alleg' ' laugh'
'high'
piano 'alleg'
on' tcdoKing| ■ (A and B laugh far
---)' and ]so we ttDto the 'eggs| 75
j" every " body ' made their contri " BUtion " j'
'from I all 'over the 'senior cdMMon
room) about |their 'point of'view about
+EGGSJ' |they were (some would 'rather
'have them tMUCH too 'soft| than |much 80
too »hard[ and |some people would
'trather not 'have an 'egg at +nALL|'
and - |s6me 'people) aim • 'thought (well)
the "thing (to do was) just | "put them in
the 'water and take them tduT again) 85
and I "then let them go on ' cooking |
with|out' 'cracking their heads| -* you
|kn6w| you got' ' |every possible' 4point
of view)*
A * (laughs)* 90 B afbout 'boiled eggs| ' (thbn| • you jwent on to the tNEXT 'topic) it was |like • as jthough there was an'un'written' a+GENda]
A '|„yeah|'- 95 B 'and |everybody 'made their contriButionj Ithen^you went on to the' • Inext 'point on the a'gendaj - and then jmary tjohnson tSAiDJ - \>H 'have a THEory) • that • aim • |one should eat| — |alTERnately| 100 jliquid and solid meals) - |so| ■ |I eat ■ tliquid at BREAK£ast| -1 have [liquid BREAKfastj - | solid LUNCH||liquid tea) ■ and [solid DiNNer) *(laughs for —) |sdMEbody* 105
f
THE LANGUAGE OF CONVERSATION 101
A *(Iaughs)*
•alleg' B I told this to| 'said she |probably'
knocks back a tbottle of g!n| for
|BREAKfaSt|
A&B (laughfor—)
'narrow* "accel"
"allegriss monot'
piaruss
'narrow' 'forte'
piamss 'piano'
'forte'
'alleg narrow'
'dimin'
'piano'
'alleg*
'narrow'
'piano'
'piano'
III
B 'wa |one of jone of thi * tGdRgonsj at saint I Paul's I "was | talking a+bout this particular school| " and saying you're " |"lucky to 'get them in cos it's a very 'difficult 'school to get'TiNto!'-
A '|„m|'
B 'and a [very tGdoo school)' • 'it's a j'BEAUtiful 'school)' • *)very NteEJ*
A *|hMTXED|* '
B |„Nd|-
A ' |niCij' • *' this |single* +sex Business)'
B *{)single) -* |bYeah) • |„yeah|---
I this s s *y°u |nKN6w| it's a • sort of—') "out" dated tpdLicy)'' which , [just 'goes 'on and 6n| ' - it |"still 'tends to be »true) that |most of the Tbest 'grammar "schools| are |single sex)' A '|„m('— "
B 'as |far as I can tGAtherj' • |best
in 'terms of — "you *|kn6w| '* A *' jrecords* *to sh6w| ' B '|„yesJ'
no
115
120
I25
I30
102
PRACTICAL ANALYSIS
piano ' piano'
'piano narrow'
piamss «high*
'alleg' "monot"
'rhythmic' ' piano'
' piano' ' lento'
piano
'high'
A 'N' 135 B ' |ry£ah] ' —
A ' of course the jl c tc seem to be
going| - |iback to this 'single tsex
„Business| |d6n't they)' - - -
*' (obscured speech) '* 140 B *' jthis* *is 'the TENDencyj' |isn't it[ -
' (I mean) I was " | talking to
'somebody on the 'phone this" tEVEning . a'bout 'this| and we were j„sXYing|' •
you I know that in tsussia| | after 145
the revotLution| * (a) |national
'policy of co educAtionj'' and |then ■
it swung tright back| ' A *|m|' — I |can't see why| be|cause ■
I'm contvinced that ■ 'tmixed 'schools 150
— 'are the |"s6uNDestj' — I |mean
tovertAix - [the |„sduNDest]| -B well it |»FEELS 'healthierJ
*|D6Esn'tit|* A *|„yes|* • 155 B and | seems ' healthierj
A  ' jnYES| '
B the |"THEOry fs| that • they dis|'tract' each 'other| — but that's |iiFE| |isn't it] 160
It does not require a very close examination to see that these extracts display certain linguistic characteristics of considerable importance, transcending whatever differences exist between them. Three factors seem to us to be central. First, there is the inexplicitness of the language, which is to a large extent due to the participants' extreme reliance for much of their information on the extra-linguistic
THE LANGUAGE OF CONVERSATION
io3
context in which the conversation is taking place. This manifests itself through the frequent use of apparent ambiguities, 'apparent' in that these are only ambiguous when isolated from their context, as on a tape. For example, there is the use of many anaphoric features of language (such as the substitute-word 'one', or the demonstratives), which produces sentences like 'That's a big one', which are unintelligible on tape without further explanation. (Tape recordings of other varieties are on the whole very explicit, and do not produce many ambiguities of this kind.) Also, there is the frequent 'incompleteness ' of many utterances, this again being but superficial, as the context makes perfectly plain to the speakers what was being intended, thus making redundant its vocal expression. There is in addition a large amount of phonologically obscure utterance on the tape, which is not due to the quality of the recording, but to the participants lowering their voices to an inaudible mumble, or to their simply tailing ofFinto silence. This is also tolerated, along with other obscurities in the course of utterance, because of the permanent possibility of recapitulation upon request by the listener, a possibility present only in certain types of dialogue, and rarely present in writing. The other aspect of conversation's inexplicitness derives from the extent to which the participants have a common personal background - in the present case, the fact that the participants knew each other well meant that they were often able to take a great deal of what they were trying to say for granted. The more one knows somebody, the more one can rely on abbreviated forms, in-slang, subtle references, family jokes, and so on. All these features of inexplicitness, which are diagnostic of conversation in the sense being discussed here, are evident throughout the extracts.
Secondly, conversation is characterised by randomness of subject-matter, and a general lack of planning. The three extracts, on the same conversational occasion, are very different: compare the relatively brief and domestic exchanges of I, the monologue on a particular theme of II, and the greater discursiveness of HI. It is not possible to predict at the beginning of a conversation how it will end, or how it will develop within any period. Conversation, as opposed to such concepts as discussion or debate, regularly lacks an overall theme. This unpredictability is of course optional. It is always possible to guide the course of a conversation towards a given theme. The point is that at any place in a conversation one may, if desired, 'change the subject' without this being felt to be linguistically inappropriate. It is
104      PRACTICAL ANALYSIS
this potential for change which is the important feature of the variety. The informality of the conversation situation is also reflected in the fact that any kind of language can occur, without its being necessarily linguistically inappropriate, including such extreme examples as complete switches in accent or dialect for humorous effect (cf the professional use of this technique by comedians), or the introduction of recognisable (albeit artificial) dialect forms to indicate familiarity or intimacy. It is significant that in an informal language situation, very formal language may be used from time to time, as in argument or humour, without its being out of place, whereas the reverse is not true. It is this juxtaposition of usually separated linguistic features which is a major characteristic of conversation. The only other variety where a comparable flexibility may be found is literature.
The third general feature of this kind of English has been regularly noted by scholars, and probably over-rated, namely, the phenomenon which has been called 'normal non-fluency'. Informal, spontaneous conversation is characterised by a very high proportion of'errors', compared with other spoken varieties, involving hesitation features of all kinds,4 slips of the tongue (though these are by no means restricted to this variety), and a substantial amount of overlapping or simultaneous speech. There are two points to bear in mind about these features, having noted their existence. First, it is not their occurrence as such which is significant, but their distribution: as has been suggested,5 hesitancy is strongly influenced by periods of creative thinking - the more one is thinking what to say, the more likely hesitation features are to appear - and this tends to produce a cyclic pattern. Secondly, and more fundamentally, even the distribution of errors has to be seen within a wider perspective. As recent discussion of the distinction between competence and performance has suggested,6 the actual occurrence of given features in a text is only one sub-set of the possible occurrences of the features in the language as a whole, and one should not pay too much attention to individual occurrences without bearing this in mind. The really significant fact about informal conversation is the toleration of these features when they occur, and indeed the expectation that they will occur. Perfect fluency in this variety tends to produce the wrong effect, for psychological and other reasons - one gets labelled a 'smooth' talker, for instance - which rather suggests that hesitation phenomena are of primary significance in determining the acceptability or otherwise of conversation. What must be avoided at all costs is prejudging this
THE LANGUAGE OF CONVERSATION I05
issue by inculcating a pejorative 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 (which is why we put our use of the word 'error' in \       inverted commas above) is in fact to judge conversation against some ;       other (usually written) standard, such as is manifested by the regular omission of these features in written forms of conversation, novels or dramatic dialogue. Considered in its own situation (that is, with gestures, facial expressions, and so on all included), conversation does not seem'disjointed'at all. \< These general points are perhaps fairly obvious. Taken along with
the less obvious and more detailed features of linguistic behaviour \ which occur in the above texts, there would seem to be very clear ? evidence that there is a valid linguistic basis for regarding this kind of ", language as a variety, in the sense in which this term was discussed in j       Chapter 3.
\ There is relatively little of significance to be noted about con-
t       versation at the phonetic level of analysis. A basic point which must r       be made is that as there are no restrictions on who may participate in i       a conversation, and as there is no formal training required, the range ;       of voice qualities one finds being used is entirely random and without j       pattern - as opposed to the more predictable qualities of certain other varieties (eg television advertising, sermons, spoken legal language). Otherwise, the only features which regularly occur (though they are not much in evidence in the texts used here) are the use of a wider range of sounds from different air-stream mechanisms and other configurations of the vocal tract than one finds in other varieties of English
> (vocalisations such as 'tut tut', various whistles (eg of amazement), artificial clearing of the throat or coughing for purposes of irony, and
> other snorts and sniffs, to communicate disgust and other attitudes), and the greater use of and permissiveness for onomatopoeic words and
[       sounds, such as ' whoosh',' boing' brrr'. 1 r The segmental phonology is also restricted in the amount of
s      stylistic distinctiveness it contains. All speakers in these extracts used *      their normal varieties of Received Pronunciation, apart from during ,       the single anecdotal excursus in IL The possibility of switching accents, already mentioned, is in evidence here, as indicated rather crudely in the margin to line 60. Frequent use is made of lexical items <t      with an abnormal syllabic structure for English, such as 'sshhh', 'mhm'. Another feature, not indicated in our transcription, is the
i
106      PRACTICAL ANALYSIS
regular use of the assimilations and elisions which have been noted as characteristic of informal English,7 and which are largely absent from many other varieties of speech, where they tend to be avoided, either for clarity of enunciation (as with the public-address system on railway stations, or in certain kinds of radio broadcast) or because of a misguided fear of being criticised as careless in articulation.
The phonological distinctiveness of conversation lies mainly in the use of non-segmental features of language. First, conversation, unlike all other spoken varieties apart from spoken literature and humour, will allow the occurrence of the entire range of prosodic and para-linguistic effects. Contrasts such as sob and cry occur relatively frequently, compared with elsewhere. There would seem to be no social restrictions on the range and depth of emotions which might be displayed in a conversational situation, the controlling factors being rather a question of personal relationship between the participants; consequently the linguistic expression of emotion, primarily a function of non-segmental features, is similarly unrestricted. In principle, then, anything may occur. In practice, within any given stretch of utterance, very little occurs; which leads us to a second main characteristic of conversation, namely, the tendency to make frequent use of a small number of basic prosodic configurations. This may be seen in a number of ways throughout the above texts. The relatively standardised, narrowed pitch contours for many of the monosyllabic response utterances provide a specific instance (22, 26, 30, 33, 44, 49, 51, etc), as does the very high proportion of simple falling tones. Another example of the tendency would be the frequency of a 'stepping down' kind of head within tone-units, that is, a sequence of gradually descending syllables from the onset to the nucleus, and the almost complete absence of a 'stepping up' type of head (where the syllables gradually rise), or heads involving wide pitch jumps between the syllables, which are common in most other spoken varieties of English. Again, while tone-units may be any length in conversation, within normal physiological limits, extending in our data from one to twenty and more words - another variation on the 'anything goes' theme - there is a strong tendency to keep them short, to break up potentially lengthy tone-units wherever possible. The average length of the units in this text is considerably shorter than that of any other variety, the vast majority falling within the range of one to five words. Moreover, a relatively high proportion of tone-units are incomplete, largely due to the nature of the interchange, often
THE LANGUAGE OF CONVERSATION 107
accompanied by reduced loudness (a ' tailing off' of the voice, eg 6, 12). Related to this is the frequent absence of end-of-utterance pauses, due to the rapid taking up of cues. (Similar effects to these last are often introduced into radio drama dialogue which tries to simulate fluency and informality of this kind.)
These are only tendencies towards uniformity, of course; any danger of a mechanical repetitiveness arising within the speech of an individual is avoided through the introduction of a large number of other prosodic contrasts at various points within the tone-unit. (There is little chance of uniformity between speakers, in view of the sub-l       stantial phonetic differences between voice qualities.) Contrastive (non-final) tonicity is extremely frequent in this variety (3, 4,6, 9, 16, 19, 23, 25, etc), and to a certain extent correlates with the high frequency of compound tones, especially the fall-plus-rise (which is particularly common in extract I). There is also the occasional completely unexpected placement of nuclear tone (54), which is only likely to happen in informal kinds of speech. The tendency noted above towards the use of simple falling tones is also varied somewhat, ['       particularly through the use of low rising-type tones on statements \       (especially in I, where the normal emotional value of these tones, i       informal friendliness or politeness, would be most appropriate), the occasional use of very emphatic tones such as the rise-fall (32, 65, 66, etc) and fall-rise (30, 42, etc), and the high proportion of narrowed tones throughout. A further means of variation is the common use of ■t       high unstressed syllables, especially in the prehead of the tone-unit ■        (24, 40, 47. 53, etc).
i_ As already mentioned, a familiar point about informal con-
versation is the frequency of silence for purposes of contrastive pause, as opposed to its being required simply for breath-taking. Voiceless I hesitation is always much more frequent than voiced in any variety f of English, but in this variety its frequency is significantly high i. (especially of the brief pause within tone-units) and it tends to occur jr relatively randomly, not just at places of major grammatical junction, [ which is more the pattern in examples of written English read aloud.8 ': Voiced hesitation is not frequent in most speakers, but what is im-■t portant is the wide range of exponence one may find for this pheno-I menon. In the present extracts, apart from the occasional 9tn, there are [ hesitant drawls (17, 20, 46), and unfinished tone-units; and in every-i day conversation one frequently hears phonetic oddities of every 1      description occurring as hesitation, such as clicks, trills, and intakes of
I08     PRACTICAL ANALYSIS
fareath. One should also note the stylistic implications of introducing hesitation phenomena of any kind into more specialised discussion (eg 12$ ff): the effect of' word-searching' helps to avoid the impression of being too knowledgeable about a topic, and builds up an alternative impression of informality. Some varieties make regular use of rhetorical tricks of this kind - for example, the cultivation of apparent spontaneity in lecturing, or in television news reporting. In the latter, the principle is often taken to extremes, pauses occurring regularly in places that normal informal conversation would rarely make use of, such as after the definite article; for example, note the pause in the second nominal group in the following:
. . . gave it to. the ARCHitect| - the • |ARCHitect. . .
As a'result of the permissible hesitation frequency in conversation, the tempo is characteristically uneven within and between utterances, though inter-pausal stretches (especially those with relatively few unstressed syllables, reflecting the absence of technical polysyllabic words) have a marked tendency towards subjective rhythmic iso-chrony (seep 36). Absolutely speaking, the speed of conversation is quite fast, but this is not the linguistically relevant point which has to be made, which is that there is no conventional pressure for conversational speed to be regular; tempo is as flexible as one wishes it to be. Linguistic contrasts in tempo do occur, of course: allegro is very frequent in II, for example. As far as other non-segmental features are concerned, one should note important variations in loudness to suggest the relative importance of what a speaker is saying. Consider the normally piano expression of phatic information (seep 121), the high forte used as anecdotal utterance initiators (see 60 ff) or at the climax of jokes. The overall prosodic build-up for a dramatic climax is in fact well illustrated by the second extract. Finally, paralinguistic features of significance in this variety are - as one might expect - the use of laugh (50, 88), spread (35), and breathy (65). .
The evidence of these extracts suggests that in conversation there is a marked tendency for non-segmental features to form a basic set of recurrent patterns, which is occasionally disturbed by the introduction of specific prosodic and paralinguistic effects. The precise nature of these patterns varies to a certain extent depending on such factors as the fluency of an individual or the modality he is using. If we compare extracts I, II, and HJ from the latter point of view, certain formal differences immediately emerge: the anecdotal charac-
[ THE LANGUAGE OF CONVERSATION 109
L" ter of II, for instance, exercises a strong influence on average tone-|; unit length. Also, level tones are frequent in H (six occasions), but are [.       completely absent from I.
• One of the interesting things which in a way helps to distinguish
I       this use of language from all others is the kind of descriptive problem !       it poses, particularly as regards the grammatical delimitation of the j       utterance and the sentence. In most other varieties, utterances are I       usually clearly definable, and sentences much more so than here. The problem is readily illustrated from extract I. The notion of utterance ;       subsumes any stretch of speech preceded and followed by a change of \       speaker (cf p 45). This suggests taking lines 23-9 as two utterances on B's part, though there are extremely cogent grammatical and semantic reasons for taking them as one. Interpolations of the kind used by A in line 26 are very significant in this variety. They are ■       usually interjectional in character; their function is primarily to indi-.       cate that attention is being maintained. We need only try taking [■      part in a conversation and withholding all such ' noises off' to prove \       their integral role - the conversation rapidly breaks down, as soon as I       our interlocutor begins to wonder whether we are really listening. However, in view of the fact that they are grammatically optional (as opposed to the m used in reply to a question, eg 32), that their distribution is governed mainly by semantic criteria, and that they could have occurred simultaneously with the interlocutor's speech, it is proposed to treat 'broken' utterances of this type (eg 23-9 and similar cases) as single utterances. We are accordingly in a position to make the descriptive generalisation that the length of utterances in this variety is much more variable than in any other variety of English. There are of course certain tendencies to adopt a given length depending on whereabouts in a conversation an utterance occurs; in the data we have examined, utterances are relatively short at the beginning, longer as topics are introduced, longer still as argument develops or an anecdote begins, and short again as the end approaches. Changes in modality and status also condition variations in utterance length. No other variety has such short utterances, telephone conversation providing an even clearer illustration of this point (see p 120).
A similar problem arises over the sentence. Before we can make any statements about relative length of sentences and relative complexity of sentence structure, we must first be clear as to our criteria for delimiting sentences from each other. This problem has already been
ft
iio     practical analysis '
mentioned in Chapter 2 (see p 45), and it exists in its most marked :' form here. Informal conversation is characterised by a large number of loosely coordinated clauses, the coordination being structurally ambiguous: it is an open question whether one takes these as sequences ? of sentences or as single compound sentences. The situation is complicated by phonetic and phonological ambiguity other than that caused ■ by the intonation: the generally rapid speed of speech and the absence of inter-clausal pause, in particular, eg
I I "am cÓLDJ and I'll |be all "right... (p 97, / 3) [ {Cf the more normal kind of problem, involving pause, illustrated by J line 16, and the completely unambiguous coordination of line 64.)
The choice of solution has implications for the stylistic analysis. Thus if we take all such sequences as separate sentences, then we can make a statement such as ' sentence length is relatively short, and in f structure displays predominantly the simple type'. On the other hand, if we take such sequences as units, then our analysis must point to a significantly high proportion of longer, more complex, and more 1 varied sentence-types. We have adopted the former solution here, on ' the grounds that it produces a simpler description. To take such sequences as single sentences would force us to make a highly complicated sub-analysis of compound, complex, and mixed sentence-types ■ (79-87 would be one type, for example), which would be of little relevance for the description of most other varieties. í
Having said this, we may now qualify the point made above by noting that if utterances do reach any substantial length, it is because of this phenomenon of loose coordination. It might be better, indeed, to refer to such a feature without using the term 'sentence' at all, ^ talking instead of 'clause-complexes'. Such a procedure would \ certainly clarify a very important point about the way in which \ conversation progresses, more in a series of loosely coordinated sentence-like structures than in a series of sharply defined sentences; but it would be a bad procedure from a stylistic point of view, as we should thereby be setting up a different grammatical theory simply to account for this variety, and this would complicate our comparative statements, producing an undesirable overlap between \ the notions of clause and sentence which we have carefully tried to       . j, avoid (see pp 46-7). ř
Other than these loosely coordinated types, sentences tend to be short. Often, a number of sentences, not separated by any kind of pause, are found within a single utterance (eg 11-12, 35-6, 46-7). j
THE LANGUAGE OF CONVERSATION iii
Minor sentences are extremely frequent, especially as response utterances (though ' response' must here be given a fairly wide definition, to include utterances which are not straight answers, such as the-noises of agreement already mentioned). Non-response minor sentences are also frequent in these extracts, particularly through the use of summarising statements (as either introductions or afterthoughts), eg 59, 119, 122. Apart from this, sentences are all of the sp(ca) type, further complexity being introduced by increasing the number of adverbials, usually in sentence-final position, eg 76 ff spcaa, 142 ff SPCAaa. One should also note the high proportion of parenthetic compound types of sentences (seep 48), in these extracts, particularly through the introduction of you know (24, 47, 87, 124, etc), though other interpolations could have been just as appropriately used (eg I mean, you see). It is probable that a modality difference conditions usage here, as such interpolations are rare in anecdote, and common in more serious types of conversation.
The use of minor sentences, along with the loose coordination discussed above, is almost certainly the basis of the impression of 'disjointedness' which many people feel is characteristic of this variety (cf our attitude to this, p 105). This is further reinforced by the absence of a stable pattern of rhythm or tempo at the phonological level. This naturally depends to a large extent on the fluency of a speaker, on his familiarity with the topic being talked about, on his experience of discussion, or joke-telling, and so on; but in view of the fact that the vast majority of conversations are not between 'conversationalists', in any 'professional' sense, it does not seem unrealistic to stress less fluent conversation as being the expected kind. Moreover it is very much to be doubted whether a more fluent norm for conversation would in fact strike people generally as being more desirable in view of the observable tendency, already noted, for too much fluency in an informal conversation to be stigmatised. The theoretical distinction between competence and performance (cf p 104) does not seem to take sufficient account of material of this kind: it is not at all clear to what extent all the non-fluencies in the above extracts can be' dismissed' as performance, in view of their frequency, their undeniable distribution in terms of regular syntactic and phonological patterns, and their clear relation to a standard behavioural response on the part of the language-user.
The disjointedness referred to, moreover, is increased by the fact that many sentences and clauses (eg 6, ?io, 24, 44, 79) are incomplete.
112      practical analysis
the language of conversation 113
This is sometimes due to a ' syntactic anacoluthon' on the part of a speaker, a re-starting of a sentence to conform more to what he • wanted to say (as in 79); but it is also fairly common for A to com- 5 plete B's sentence, or vice versa (eg 133), or for the two speakers to ■ provide an ending for a sentence simultaneously - a situation frequently parodied by comedians. Once again, the linguistic point to be made is that, whether such things happen or not in any given piece of conversation, they could have taken place without being felt to be inappropriate. We should also note in this connection the way in which the characteristic ' give and take' of a successful conversation is \ maintained (see extract HI, in particular, where things have really ' warmed up'): the pace of the dialogue is kept up by the' agreement' ' question-tags and the phatic interpolations. In a dialogue of this type J it is sometimes difficult to find an obvious stopping-point for extract- 1 ing a linguistic sample: in the present case, the cut-off in HI is quite arbitrary, as the topic continued to be discussed for some minutes. ;
There are a few other points to be noted at the level of sentence • and above. Overt, inter-sentence linkage is very marked: the extracts provide illustrations of all the types referred to on p 44 (personal     ' 1 pronoun reference, cross-reference using the articles and determiners, I ellipsis, etc). Interrogative sentence types are particularly frequent. Imperatives are few, and when they do occur, their force is' softened' ; through some device (eg the additional clause in 9), as better befits the ; informality of the situation. Finally at this level, one should note the frequency with which speakers make use of different grammatical modes of reference, such as reported speech, directly quoted utterance (both of these especially in H), and undefinable mixtures, such as the structure in line 27.
There is little to say about clause structure. Vocatives are common, especially in initial position, though this is not well represented in the above extracts, presumably because the identifying or attention-getting function of the vocative is not likely to be frequently used in '« conversations involving only two people. Nominal groups tend to be infrequent as subject; the personal pronoun is more in evidence -especially the first person, which is an expected, but nonetheless a distinctive feature of conversation. One might also note in this connection the use of the informal you (71), in its impersonal function as against the more formal one or in place of the third person pronoun (91).
Group structure, both nominal and verbal, is relatively uncompli- | cated. The former tends to be of the simple type Determiner (Adjective) j
Noun, with little postmodification or adjective sequence. There is a tendency for nominal group structure to increase in complexity as the level of seriousness of the conversation increases (cf I and HI from this point of view). Within the nominal group, one should note the frequent use of a very limited range of adverbial intensifiers such as very, a bit (contrast lecturing, with its use of highly, notably, etc); relative clauses usually omit an optional relative pronoun (eg 105); it is normal to put the preposition in a relative clause at the end (eg 107)..
Verbal groups are also simple in structure, usually one auxiliary with a lexical verb, though the whole range of auxiliary combinations is possible. A significant flexibility is that conversation allows the occurrence of the whole range of tense-forms and aspects. Also highly distinctive is the occurrence of contracted verbal forms (he's, I'll, etc); the frequency of informal 'filler' verbs, such 2&got (5, 16, 88, etc); the tendency to use phrasal verbs (probably below conversational average in the above extracts); the infrequency of passives; and the use of colloquial ellipses (eg 1, 62-4).
Finally, at the grammatical level, informal conversation provides the best example of a variety wherein points of disputed usage tend to be passed over unnoticed. It is perfectly possible to hear two people continually using alternative constructions in the same conversation, neither noticing the difference introduced by the other, and often a person will be markedly inconsistent himself, without this being noticed. They only become sensitive to points of usage in a relatively self-conscious or formal situation, when pressures to use a particular form and reject another on some obscure ground of'correctness' regularly come to the fore. As a result, one is liable to find in conversation instances of both a favoured and a condemned form (by the standards of traditional grammar books). For example, in 76 one finds everybody made their contribution. The linguistic point which has to be made immediately is not that their is wrong and his right, or"vice versa, but that in this variety either is permissible, and will be used depending on one's linguistic background. In normal conversation, no comment would be made on the choice of usage: only the pedant, whom one trusts to be exceptional, tries to introduce ' linguistic disharmony * into this variety. Similarly, repetitious structures,.looseness of syntax, 'weak' words like got and nice, and so on - all of which would be condemned, and with good reason, in children's school essays and elsewhere - are a standard and indeed a valuable part of informal conversation. Formal written English and informal spoken
114     practical analysis
English are two very different varieties, and the criteria of acceptable usage must not be confused.
Probably the most noticeable aspect of informal conversation is its vocabulary. Words tend to be very simple in structure: even when discussion is well under way (as in UT), there is an avoidance of specialised terms and formal phraseology, and whenever they are used, their force is usually played down by the speaker, through the use of hesitation, or the use of you know, sort of, and so on (eg 124). The lack of precision in such matters of word-selection does not seem to matter; inexplicit references are accepted (eg 112); and it is even possible to replace a lexical item by a completely non-specific prop word, as in the use of thingummy, what-do-you-call-it, you-know-tvhat-l-mean, etc, which may all function as nouns. On the whole, the vocabulary reflects the relative domesticity of the subject matter - knitting, education, table, the weather - with the addition of a great deal of phatic (' atmosphere-setting') vocabulary, and the vocalisations (such as m) which keep the conversation going.
The informality of this text is evident throughout, particularly because of its readiness to use certain items which are highly frequent in conversational English, eg yeah (44, 49, 51, etc), cos (2, 29, 114), got (1, 5, 13, 16, etc), all right (3, 11), just (2, 12, 21, etc), a bit (2, 28), fed up (51), sort of (124), sniffy (2), I mean (142), knocks back (108), warmed up (4), take to (45), a lot (54), and phone (143). One finds in the extracts a representative number of colloquial idioms, such as in a minute (11), just couldn't face (21), the simple truth is (65), the thing to do (84), as far as I can gather (131); and also the occasional cliche, such as that's life (160) - though again this does not have the undesirable connotations that cliches have in certain other varieties. Informal conversation is also characterised by a great deal of lexical hyperbole, usually with phonological support, eg stupid (14), five thousand (14), freezing (32), every possible (88), on and on (126), all over (77). 'In-group' slang is also frequent - for example, the use of abbreviations familiar to both participants, such as St Paul's (112), LCC (137). One should also note the tendency, illustrated in the piece of telephone conversation below, to use such phrases as ie, eg, a, b, c, in a speaker who has been educated to a reasonably high degree. Familiar euphemisms can also be expected to occur - tiny garment (36) is in fact a humorous query as to the possibility of A having become pregnant. Finally, one should note the deliberate introduction of incongruous or humorous items of vocabulary into an informal conversation, the nature of the wit
1 f 1
i
the language of conversation 115
; of course largely depending on the common background of the 1 speakers. In the present case, we have human being (s6),gorgons (ill), • and possibly eat liquid (102) and string (46). Related to this is the introduction of vocabulary normally part of another province or j. idiolect for dramatic or humorous effect, as with communally (67), principal (61), and did (75).
Semantically, the most important feature of this variety is the 1       randomness of the subject matter, the lack of an overall contrived :       pattern, the absence of any conscious planning as conversation pro-:       ceeds. Conversation does not take place in a series of coordinated blocks, but - especially as someone searches for the beginning of a ;       topic - in a series of jumps (as in extract I). There is a general absence of linguistic or cultural pressures to make the conversation go in a particular direction,9 and there is a corresponding admission of all I       kinds of spontaneous effect, especially switches in modality. Many features are indicative of this: the simultaneous start given to an utterance (eg 33-4), A supplying B's image (133), the occurrence of ;       afterthoughts (n&ff, I53#), the loose stringing together of ideas ■       (9°ff> I4I#)» the rough synonymy (118 ff), the repetitive nature of >       certain parts of the discourse (such as the multiple agreement in 134-6), \       and the redundancy which allows omissions (24). Other important , I       semantic points have already been mentioned: the freedom to intro-"       duce material of almost any kind (the limits depending on the sex, class, and intimacy of the participants), such as jokes, bathos (eg 61 ff), irony (64 - cf And Sir said . . .), and accent- or dialect-switching (as in ;       60 ff);10 the importance of intimacy-signals, silence-fillers,' rapport-'< ;       makers', or whatever one calls them; and the importance of the context in which the utterance took place, so that omissions go unnoticed (24), speech which is obscure to the analyst is understood by the ;       participants (10), and so on.
To call the language used in the above situation a ' variety' is ] ; perhaps a little premature. The term'informal' is readily correctable "s ;, with certain linguistic variables, operating at all levels, and reflecting 1 ; the parity of social status of the participants and the spontaneity of ' their expression. The term' conversation' is not so clearly distinguishable from other terms which come to mind, in particular the notion l of'discussion'. It may well be that a useful linguistic distinction can i ; be drawn between 'conversation* and 'discussion', in terms of the i i degree of seriousness of the subject matter, or the formality of the i \      occasion, and it is not difficult to think up probable linguistic corre-
u6
PRACTICAL ANALYSIS
lates - for example, at the semantic level, the relatively monothematic nature of discussion would condition a markedly different semantic structure for the discourse from that which exists in conversation. But it is unlikely that there is a clear boundary between conversation and discussion: there are many intuitively clear stylistic categories within which elements of both conversation and discussion combine - the concept of' talking shop', for example. As a consequence, we do not wish to suggest that any clear lines of stylistic demarcation have been drawn in the present instance. We have tried to point at the centre of a peripherally unclear stylistic issue; we feel that there is sufficient evidence to make the postulation of a variety warrantable; but there is much which remains to be investigated.
We shall conclude this chapter by examining one outstandingly neglected area in greater detail, namely, the question of how far other kinds of informal conversation occurring in a more restricted situation may best be analysed: whether they should be described as instances (more precisely, sub-provinces) of the same province, or as separate provinces. The answer will of course depend on the extent to which these more restricted varieties of conversation share the properties of the variety which we have described so far. If there is a very close linguistic similarity, the former solution will be preferable; if there is little in common, then the latter, and a label other than ' conversation' will have to be found.11 We have chosen an extract of telephone conversation to illustrate this problem.
IV
A jhighview double three four five)
B good |MORNingj
A *(hel|L6[ |ARthur|)*
B *rvALerie|*
A |yes| jgood JMdRNingif
B tthi thisf is JARthur SPEAKingj
A hel|Ld|
B 3 |soRsy I've 'been so 'long in 'getting in t6uch 'with you| I
:
'high'
'alleg*
'alleg' "monot' 'rail' 'alleg'
'alleg'
* alleg high narrow'
'alleg'
'creak*
THE LANGUAGE OF CONVERSATION
117
io
I rang a tcduple of times YESterdayj
and you [weren't in| A |bn6| I was in |c&LLege 'yesterday) B you jwere| A |yes| *and I*
B *jaHA|* 15 A (thought that might «HAPPen|
'but jnot to w6RRy| • (what T
wanted' to say to you tREALLyj was
am - ' 11 didn't know 'whether you
were going to say' • that you tcould 20
come or you tcduLDn'tj' but I was
"|going to say 'could you 'make it" '
'the tF6LLOwing 'Saturday)' • B a:m ' |„yes| jwell' - 6ne| |I was tgoing
to 'say that I • that we fwERE c6Mingj 25
A |yes| • *|spl£Ndid|*
B *|and „tw6|* • we ' |can make it the
'following 'SaturdayI' A |nCAN you| |only om it's it's a tMlNor
complicAtion) but am 30 B ' js6RRy| |didn't get "that|' A the I point is that my tcHiLdren| -
are | going away for the tweekENDJ -B |yes|
A 'and it was |going to be tnTHis 35
weekend| and jnow it's going to be
+next| *and* B *'|oh'*
A it's jreally more comvEnient for me| if they're [not here] bej cause f otherwise I thave to 40 keep tflapping aROUND|' and
IIS
'low creak'
•low*
'alleg* "high"
' acce)'
' low creak' 'high'
'alleg* ' high'
' acce)'
PRACTICAL ANALYSIS B mJHM|
A [DEAiing with them| *'you* [kn6w| -B *'Jyep|'*
A |so am • we'll 'make it the tF6LLowing
nSATurday *thenj* B *' that's* |^ne| |yesj' |sameTiME| A [same time| (yes) B |g6od|
A am ' " [do 'you think" - |I don't even know twhich • I j can't even re'member what the chap's +name is|' the |other chap in your dePARTment| • |BERnard is it]
B [bernard blfl am: • [green'field) ■ A yeah |not BL6oMfield| (laughs) B !„yeah| •
A 'so j could you 'mention it to Hiiw)
cos |I've in'vited him as well) ' B '|YEs)'--*'|ote|'*
A *|ok|* • '|i^ne| |everything 'all ri'ght|' B |öhj jriNEJ' |was 'there 'anything' else) am:
A |nö| I ijust a I've jleft some tRECords) in |smart's room last 'night| 'which I was all |pANic 'stricken a'bout) cos they're jnot m1ne| ' ■
B |m| •
A but I jtold neil| and I |hope he am ■ got the pöint| so I *\jmt wanted to*
B *a I [don't* * I'm |not sure whether he 'quite 'got - the MESSagej [would you
45
50
55
60
65
70
'alleg'
'alleg'
'alleg'
THE LANGUAGE OF CONVERSATION IIO
'tell me a+GAiN 'please| A |yes| |there's aim • a tRECords) in
I smart's r6om| it's | measure for 75
MEAsure) • |in [in an ALbumj — B |yes| •
A and am I |left them last night| by mis|TAKE|
B |Kihm| 80 A and they're jnot mine) so ' |that
means I've 'got to take 'special
tnCARE of them| and I jwant to
collect them to+M6RRow|' • B |„yes| do you [want me to 'get h6ld 85
of them -for *you|* A *' |could* you just 'put them* somewhere
CAREfully fFOR meft B t|Put them1}- safe| •
A *)„yIahj* 90
B *|yes|*- I'll |d6 that)
A ' [thanks very much 'arthur|'
The telephone situation is quite unique, being the only frequently occurring case of a conversation in which the participants (and of course the contexts in which they speak) are not visible to each other. As a result, certain differences between this kind of conversation and that already described become immediately apparent. A different range of situational pressures is exerted upon the participants, and consequently the range of linguistic contrasts which they are permitted to choose differs somewhat. They cannot rely on the extra-linguistic context to resolve ambiguities in speech (such as in the use of ambiguous demonstratives, pronouns, etc); visual feedback being absent, auditory cues become all-important, and in view of the diminished quality of the voice over the telephone, there develops a greater uncertainty and confusion in maintaining the ' give and take' of the dialogue (which was rarely impeded in extracts I, II, and ID);
120      PRACTICAL ANALYSIS
there is a strong pressure for greater explicitness, arising out of the quality of the medium of transmission, for example having to spell out words because of the distortion of certain sounds; and there is a tendency to avoid long utterances without introducing pauses which allow one's listener to confirm his continued interest, and his continued auditory'presence'.
This last point is worth developing. The phonological system of pause that we make use of in English varies to a certain extent from variety to variety, and telephone conversation represents one extreme. Here the total number of contrasts available to most varieties is much reduced. We cannot make use of the longer pausal contrasts, because anything approaching a silence on the part of one of the speakers is either interpreted as a breakdown of communication (Hello ? Are you there ?) or as an opportunity for interruption which may not have been desired. This is particularly the case when such a conversation has not progressed onto a set theme: there is either a complete absence of pause, especially between sentences (eg 9, 12, 21, 24, 29), or a brief pause (eg 17). Longer pauses are usually restricted to grammatical contexts which are clearly incomplete (and which are therefore liable to be uninterrupted), and frequently reinforced by voiced hesitation (eg 19, 27). There is a tendency not to be silent before answering a question or introducing a new topic: if a delay is required, then voiced hesitation is usually introduced to 'fill the gap' (eg 24, 45, 50). As a result, voiced hesitation of different kinds (eg drawls, random vocalisations, repetitions of words) is proportionately more frequent here than elsewhere. The silent pause system, also, is reduced to a basic three terms, zero, brief (•) and unit (-), with double (—) occasionally being used (cf p 35). There is nothing longer. We may contrast this with the opposite extreme, where maximum use is made of pausal contrasts, namely in the language of public speaking, where, in addition to the above, treble, and even longer pauses are possible. When there is no one to interrupt, a speaker can manipulate silence freely. (Interestingly, some of the best ripostes during a political speech come at a point when the speaker is trying to gain maximum effect through a rhetorical silence: it is easy for an inexperienced speaker to be thrown completely offbalance by a punctured silence.)
There are a few other differences between telephone and other informal conversation. In view of the purpose of a telephone call, questions, responses, and imperatives are all likely to be frequent. Again, the purpose of a telephone call in the majority of cases implies
J THE LANGUAGE OF CONVERSATION 121
t        a specific theme, or set of themes, which have to be raised, and this I1        has implications for the semantic structure of the discourse. And there \        are undoubtedly some minor points which a full description would (        have to cover, for example, the different kind of formulaicness at the opening and close, or the different senses of hello (meaning ' I am here' or ' Are you there?' rather than simply 'Greetings'). But apart ;        from this, it is difficult to suggest any linguistic features that could not equally well have turned up in the earlier passages of conversation.
* There is the same listing of dominant features at sentence, clause, and group levels, for example; the same descriptive problems emerge (cf
I the loose coordination of 78 ff, for example); in vocabulary there is ^ the same use of colloquialism, idiom, and vocalisation, apart from the minor differences noted above. In other words, it can be argued that while the range of variety markers is considerably diminished in tele-J phone conversation (as compared with I, II, and III), the kind of » marker which occurs (with the one exception of the distinctive pausal i       system) is essentially the same. The conclusion which suggests itself,
• therefore, is that telephone conversation and other conversation are > different only in degree, and that the former can most realistically be \       seen as a sub-province of the more general notion.
[ Exercises
i i Examine the markers of informality in the extracts and decide ' which have the most important stylistic function.
2 The exttacts provide a clear example of the introductory, ice-breaking use of language known as 'phatic communion' (for a
, further discussion of which, see R. Quirk, The Use of English,
; J Longmans, 1962, Chapter 4). But there is more to phatic com-
! ; munion in English than talk about the weather. To what extent
5 [ does the kind of phatic communion vary depending on differences
( in (a) province, (b) status, and (c) dialect ?
3 What other kinds of modality difference regularly occur in conversation?
4 In what ways does radio drama dialogue differ from the dialogue j described in this chapter ?
I ^ 5 The following is an extract from Everything in the Garden by Giles j ■ Cooper. In what ways does this conversation differ from the kind
illustrated in this chapter ? What spoken information is left out of j , the written version ?
122     PRACTICAL ANALYSIS
Jenny: When do you want to eat? Bernard: When I'm fed.
Jenny: No you won't, you say that but you never do and then it all
gets cold while you finish something. Bernard: What is it ? Jenny: What would you like ? Bernard: What is there? Jenny: Nothing much. Bernard: Then I'll have it cold with pickles.
Jenny: There isn't time, if I cooked it now it would take two hours
and it wouldn't be cold till midnight. Bernard: It was only a joke. Jenny: I thought you meant the joint.
Bernard: No, you said nothing much, and I said I'd have it cold
with pickles. Jenny: I'm not there.
Bernard: Because you said what would I have and I said what was
there, and you said ... Jenny (breaking in): All right, eggs ?
6 How does the novelist try to reflect conversation? Discuss the linguistic features of the following extract, paying special attention to the way in which the author provides us with clues as to how the speech of the characters should be interpreted.
' Why don't you have a bicycle, and go out on it?' Arthur was saying.
' But I can't ride,' said Alvina.
' You'd learn in a couple of lessons. There's nothing in riding a bicycle.'
' I don't believe I ever should,' laughed Alvina. ' You don't mean to say you're nervous ?' said Arthur rudely and sneeringly. 'I am,'she persisted.
' You needn't be nervous with me,' smiled Albert broadly, with his odd, genuine gallantry. 'I'll hold you on.'
' But I haven't got a bicycle,' said Alvina, feeling she was slowly colouring to a deep, uneasy blush.
'You can have mine to learn on,' said Lottie. 'Albert will look after it.'
r
>
THE LANGUAGE OF CONVERSATION 123
' There's your chance,' said Arthur rudely. 'Take it while you've got it.' (D- H. Lawrence, The Lost Girl*)
* Copyright 1921 by Thomas B. Seltzer, 1949 by Frieda Lawrence. Reprinted by permission of the author's agents, the Estate of the late Mrs Frieda Lawrence, and The Viking Press, Inc.
Notes
1 One might well find that for more restricted studies of English varieties a different yardstick would be more useful; for example, someone making a comparative study of written varieties might find it more valuable to choose a written variety as a basis for investigation; or a study of types of public speaking might be more usefully undertaken if a more formal variety of spoken English were chosen to begin with.
2 d. abercrombie, 'Conversation and Spoken Prose', English Language Teaching, 18, 1963, pp 10-id; a. h. smith and r. quirk, 'Some Problems of Verbal Communication', Transactions of the Yorkshire Dialect Society, 9, 1955, pp 10-20.
3 Naturally, permission was asked to make use of the recording, and all participants agreed. Further, to ensure complete anonymity, all names were altered to rhythmically identical equivalents, a procedure which we also use, for obvious reasons, in the extract of spoken legal language in Chapter 9.
4 These include phonological, grammatical, and lexical types, eg a higher proportion of anacolutha and word parrials alongside the familiar 'ers' and 'urns'. Cf j. BLANKENSHip and c. Kay, 'Hesitation Phenomena in English Speech: a Study in Distribution', Word, 20, 1964, pp 360-72.
5 For example, by f. goldman-eisler, 'Sequential Patterns and Cognitive Processes in Speech', Language and Speech, 10, pp 122-32.
6 See n. chomsky, Aspects of the Theory of Syntax, M.I.T. Press, 196s, p if
7 For example, by a. c. gimson, introduction to the Pronunciation of English, Arnold, 1962, pp 26) ff.
8 It would be interesting to see how far silence in conversation was being supplemented at any given point by an overt reliance on context (eg by some bodily gesture, as when one finishes a sentence with a shrug of the shoulders instead of a word), but only surreptitiously filmed material will do this adequately, and expense makes research difficult.
9 Those cases where X comes to talk about a particular subject with Y would not be included by this statement; but these would tend not to be informal in character and would in any case involve a certain amount of random ' beating about the bush" before the participants got down to business. The only genuine exception which we can think of is the 'angry scene* conversation, which usually mixes informality with formality, where X wishes to get something straight with Y without further ado.
10 Here, one should note that it is necessary only to begin well, in imitating someone's speech informally, and to give an occasional reminder that one is still imitating. Absolute consistency is unnecessary (except in professional circumstances, of course, and even there few narrators are perfect): in the present
124     PRACTICAL ANALYSIS
text B slips out of the Principal's accent quite quickly, and introduces the occasional grammatical change, eg the use of have to (71), where the Principal would probably have used mast. 11 If one wished, the 'similarity' between the two texts could be quantified statistically (cf p 22 above). There is a finite number of linguistic parameters recognised in the description, and these are ordered in a given way; consequently it would not be difficult to arrive at an overall statistic which would characterise a text. One could plot degrees of similarity using standard techniq ues. The only problem would come when a text could be shown to fall perfectly in between two such extremes as conversation and discussion. In such a case (which we have not yet come across), one would have to postulate a new stylistic category, rather than force the text into either extreme.
Chapter $
The Language of Unscripted Commentary
Most commentaries have something to do with description, explanation, or opinion. But the three are not always present in equal proportions. Some forms of written commentary, for instance, by providing the supplementary information which will enable a text to be more fully understood, set out purely to explain. In spoken commentary, on the other hand, the need for vivid description is often so strong as to reduce explanation to a minimum. And commentaries are to be found - notably of the political kind, both spoken and written - in which there is a great deal of opinion but precious little that is either described or explained. If it is remembered that the descriptions, explanations and opinions may, on different occasions, relate to an almost unlimited range of subject matter, it becomes obvious that the term 'commentary' has to serve for many kinds of linguistic activity, all of which would need to be represented in any adequate descriptive treatment, and would presumably require separate labels such as 'exegesis', 'political comment', and so on.
The aim of this chapter, however, is not to compile an exhaustive list of all the imaginable types, but to discuss one or two examples of what is meant by 'commentary* when the word is used in its commonest current sense. There is little doubt that for most people nowadays a commentary is a spoken account of events which are actually taking place, given for the benefit of listeners who cannot see them. There are of course many occasions when both commentator and listener are looking at the same event - notably on television -but here the activity is usually self-evident and most commentators are mercifully aware of the absurdity, or even impertinence, of reporting that the ball is in the net, the stumps are spreadeagled or the parade commander has fallen from his horse. In other words, the television commentator's most useful function is to provide background information or explain any bits of activity that do not explain themselves. In contexts where the audience cannot see the event the