[To make the “prosodic” graphics clearer and more compact, I have used Times New Roman font, whilst the main text, aside from grammatical terms, is still in Arial.]
VI.A Four units
VI.1 The domains of PROSODY and GRAMMAR are more elaborately related than most “grammar-books” suggest. To start us off, we can distinguish four units in “real English”. The SENTENCE is the preferred unit of written English; it is realised by means of CLAUSES; and it has been the unquestioned central and perhaps “largest unit” of many “grammar-books”. The UTTERANCE is the preferred unit of spoken English; it is realised by means of TONE GROUPS; and should not be used interchangeably with SENTENCE.
VI.2 These distinctions have not been conscientiously maintained in the past. “Grammar” has been tacitly thought to apply best to written English — the term even came from Greek “gramma” meaning “letter”. Spoken English seems, justly I think, to involve considerably greater personal, regional, and social variation, which is mistaken for instability or even unreliability, and for an impediment to “standardisation” (II.16). Moreover, even at today’s levels of technology, speech is laborious to observe, record, and transcribe, much less make over into visual displays for circulation. If memory serves, Manny Schegloff, of “conversation(al) analysis” fame, once told me that one recorded hour takes 20 hours to transcribe with the system his group was using, which offers impressive detail about timing, overlaps, Californian laid-back pronunciation, and so on.
VI.3 To help restore the balance, we can explore GRAMMAR together with PROSODY — also called “intonation” in various works [Note 1] — that is, the overall organisation of Spoken English and, I believe, also mentally applied while writing: I at least never write and finalise anything for public use without sounding it out in my head — advice I seized upon years ago from the stalwart Leslie Fiedler.
VI.4 Many phenomena in the real GRAMMAR cannot be sensibly accounted for any other way. Yet even the massive Longman Grammar of Written and Spoken English, despite the title, leaves PROSODY out of the picture, reassuring us that “for many purposes of grammatical research, the absence of prosodic information may make comparatively little difference”.[Note 2]
VI.6 So I am back to relying on the PROSODY of my Englishes, of which I can produce a number or varieties acquired in my mobile existence. Having thus grown keenly aware of the remarkable range of personal and regional variations, I can only hope that the issues raised here enjoy some degree of general relevance. I would be happy to get data about your prosodies, which I may someday be able to report within an updated version.
VI.7 At least, I ought to preface the presentation proper by highlighting the spontaneous nature of spoken English in real conversation. There, the PROSODY and GRAMMAR manifest some classes of events, shown here in BNC transcriptions, that generally escape attention and are absent from conventional representations in written English. In a RESTART, a PATTERN is begun more than once [1-3]. In a REPLAY, a whole PATTERN re-occurs [4-6].
[1] I think there’s I think there’s quite a few
[2] if you, if you, if you want to get into it you might find you want to do them all
[3] You play two leagues, were they, were they two leagues? when I, when I was playing pool we played in the Tuesday league and the Thursday league
[4] How many questions are you gonna have in it Dad? How many questions are you gonna have in your quiz?
[5] I’m a bit torn cos I in a way that’s what I ought to be doing even though the welfare business in a way that’s wh that’s what I ought to be doing
[6] I didn’t send a message, no I’m sure I didn’t send a message chasing them
In a SHIFT, a PATTERN is begun but then abandoned for another [7-9]. In a CUTOFF, the PATTERN is entirely abandoned, and the speaker trails off or gets interrupted [10-12].
[7] me and this other mate he’d a, did a, I can, I can remember him saying something about I’ll race yuh
[8] It’s just a very good way of er teaching chi- I mean children will I mean that’s the kind of experience
[9] got a couple of oak trees, there in the corner you’ve got, ‘cos our fence you’ve got that, my house is there right, road down like that
[10] “People have got a, yeah” “Just close enough to sti still remember a lot of the cloths and stuff.”
[11] “But my bloody green jumper's all got” “Mm.” “and I've never washed it”
[12] “he was the caretaker at father's school, wonderful man, absolutely, he's got god knows what he hasn't got wrong with him, but he has sort of” “which he's made himself” “Good lord”
In a BLEND, competing PATTERNS get mixed [13-15]. In a REPAIR, an undesired choice gets replaced by a desired one [16-18].
[13] I can’t avoid not being at that governor’s meeting [avoid being + not be]
[14] I couldn’t help from crying [help crying + keep from crying]
[15] you’ve got a fencing like that and with got trees all around it. [with trees + it has got trees]
[16] I can take her up to Bristol er Cardiff next week
[17] we’d all sit down and do nothing while the old erm, the mod people had obviously it
[18] had the top been specially glazed ehm, ehm, what I mean, barred?
Some speakers are egregiously prone to these events, such as Dumbya Bush: [Note 3]
[19] This is a — any strike’s a tough — tough situation, but this one happens to come at a — or a lockout is a tough situation, or no work is a tough situation — is to come at bad time.
[20] I hope investors, you know — secondly, I hope investors hold investments for periods of time — that I’ve always found the best investments are those that you salt away based on economics.
[21] There’s what they call “actionable intelligence”, to which our military has responded on a quick basis is improving.
[22] the person who runs FEMA [Federal Emergency Management Agency] is the first voice, often times, of someone whose life has been turned upside down hears from.
His supporters react to Dumbya like her “hearers” to mad Ophelia: his “speech is nothing, yet the unshaped use of it doth move the hearers to collection; they aim at it, and botch the words up fit to their own thoughts”, which “make one think there might be thought, though nothing sure, yet much unhappily” (Hamlet IV, v, 7-12).
VI.8 Written texts represent such incongruities mainly for explicit purposes, e.g.:
[23] She hid her face from him. Mungo was embarrassed. “I…I’m…There’s no need to…” he stammered. There was a pause. (Forest of the Night)
[24] “Do you want to go?” “No. I mean, yes. No, I —” his voice faltered. Then broke. […] “Is it — about” she hesitated. “Is it about — your mother?” (Diamond Waterfall)
VI.9 For this presentation, PROSODY can be described as the acoustic shape and flow conferred upon spoken discourse by the voice, whether actually uttered in speaking and hearing, or mentally perceived in writing and reading. The ancient Greeks already grasped its vital importance: [Note 4]
[25] Skill consists in employing the voice for each emotion, [and the] speech rhythms for each subject-matter. (Aristotle, Rhetoric)
Modern studies of language, as I have noted, have preferred to leave PROSODY on the margin, except for occasional wide-ranging “functional” pioneers like Dwight Bolinger,[Note 5] to whom book is gratefully dedicated. Speech was construed as a subject-matter for phonetics and phonology, which correspond roughly to “naming the parts of speech”: find and label the pieces without sufficient attention to integrated combinations. “Applied linguistics” has until recently taken much the same route, such that we face generations of foreign learners meticulously pronouncing sounds, yet speaking with an unnatural PROSODY reminiscent of early computer speech synthesizers.
VI.10 In my conception, PROSODY implicates four scalar parameters. STRESS ranges between stronger and weaker; PITCH ranges between higher and lower; VOLUME ranges between louder and softer; and PACE ranges between slower and faster.[Note 6] A value on one parameter may describe a particular manner of speaking, perhaps to evoke the speaker’s condition or intention, such as “high” for defiance [26], “low” for condolence [27], “loud” for boldness [28], “soft” for disappointment [29], “slow” for reminiscence [30], and “fast” for belligerence [31].
[26] “Deceit is not my fault!” I cried out in a savage, high voice. (Eyre)
[27] “It’s Phipps”, the constable said, in a low voice. “He’s dead” (English Crime)
[28] She said loudly, to cover the noise of her heart, “I’m not frightened.” (Carrie’s War)
[29] His voice faded away. […] “Oh”, he said softly. “So Rachel was right.” (Edge)
[30] “I’ve read many stories of people taken but not returned”, the Bookman said slowly, as if thinking of a distant memory. (Endill)
[31] “I am going through with that [inquiry]. Only” — and there he spoke a little faster — “I won’t let any man call me names outside this court”. (Lord Jim)
In authentic speech, these four parameters naturally interact with each other, which may well explain why they are rarely distinguished in research or instruction.
VI.11 To keep the description within bounds, I shall chiefly deploy the term STRESS with three degrees. Where useful, I set an upright line | for a PAUSE and two | | for a more decisive PAUSE; these may be variously indicated by silence or just an audibly lower PITCH or slower PACE (or both). A raised dot · can handily to indicate the bounds between SYLLABLES, though STRESS can affect neighbouring SYLLABLES as well.[Note 7] Consistent with my uses so far, STRONG STRESS, shown here with a raised mark !, is articulated with the most force, and so tends to be higher, louder, and slower as well. WEAK STRESS, shown with a lowered and inverted mark ¡, is articulated with less force and tends to be lower, softer, and faster as well. UNSTRESSED, shown with no mark, is articulated with the least force. Naturally, these three levels of STRESS are not absolute, but mutually relative within the speech contour of a particular speaker, utterance, or language variety; except for the inconvenience, we might say “stronger”, “weaker”, and “least stress”. For example, Ironess Thatcher’s personal “warrior” speech contour might qualify her as Hurricane Maggie (only far more destructive in the long term) in comparison, say, to the silky gentility of Jeremy Irons.
VI.12 If PROSODY is sensitive to individual interpretation, my demonstrations all carry the reservation of representing my own performance (VI.5). Here are two of my instances, namely the renowned opening lines from Shakespeare’s Henry V [32] and Richard III [33] with the STRESSES as shown:
[32] ¡Oh, for a ¡Muse of !Fire | to as·¡cend the ¡bright·est !heav·en of in·!ven·tion
[33] ¡Now is the !win·ter of our ¡dis·con·!tent | ¡made !glor·i·ous !sum·mer by this !sun of ¡York
The STRONG STRESSES fall where I believe the most WEIGHT is merited. In [32], “muses are rarely on “fire”; “heaven” should radiate the greatest “brightness”; and “invention” is the actual power being implored. In [33], “winter” and “summer” can be STRESSED for contrast; and the “sun” is the key; whereas the audience are expected to know that the house of “York” has been secured at this turn of the War of the Roses, which Shakespeare himself had dramatized in the earlier play Henry VI. Also, I would briefly hesitate after “Fire” and “discontent” to equalize the BALANCE of the two PATTERNS.
VI.13 Types of STRESS can be more dynamically displayed in their interactions with PITCH and PACE, using arrows as visual aids. CERTAIN STRESS, typical for STATEMENTS, has a slow fall and can be either STRONG if spoken with emphasis [34] or else WEAK without it [35]. UNCERTAIN STRESS, used in some types of QUESTIONS, has a slow rise, either STRONG for excited [36], or WEAK for calm [37]. Both CERTAIN and UNCERTAIN may also have the stressed SYLLABLE at a slower Pace than the unstressed ones. SPIKED STRESS, typical of short, sharp INTERJECTIONS and COMMANDS, has a fast fall [38].
For some speakers like myself, and in fairly isolated STRESS positions, a subtle and brief compensatory movement may occur, drawn here as dotted arrows as distinct from solid arrows for the principal movements. What goes down must first come up, and vice-versa.
VI.14 Minor variations with the slowest PACE include DELIBERATIVE STRESS for “deliberating” what to say, with a long fall and long rise [39-40]; and EVALUATIVE STRESS, when “evaluating” what’s good or bad, with a long rise and long fall [41-42].
These two variations, which some speakers may consider overly weighted, are especially prone to spread across several SYLLABLES.
VI.15 In English, a WORD spoken by itself and having more than one SYLLABLE normally shows STRESS on at least one [43]. With four or more SYLLABLES, STRONG STRESS often goes to one and WEAK STRESS to another [44]; or SYLLABLES get compressed to make fewer, as in !in·ter·¡est·ing => !in·trest·ing (III.21). A PHRASE can assign STRONG STRESS to a key item near the end for END WEIGHT (cf. § XXX), e.g., the NOUN in a NOUN PHRASE [45] or the PROCESS VERB (i.e. the one telling the ACTIONS or EVENTS in a VERB PHRASE in [46], with only WEAK STRESS elsewhere. Yet STRONG STRESS may be deferred if the PHRASE continues, such as with a POST-MODIFIER of the NOUN in [47], or an ADVERB of the PROCESS VERB (the one in [48], whilst the NOUN or VERB just gets WEAK STRESS.
[43] “Was I very bad?” “!Aw·ful”. (Ruddigore)
[44] EN·¡VI·RON·!MENT·AL·ISTS. Friends of the Earth propose green taxes (Autocar)
[45] An ex·¡treme·ly at·¡tract·ive !plant. (Aquarium Plants)
[46] Until she was sixteen they had been ¡con·stantl·y !mov·ing (Healing Fire)
[47] I particularly noticed one ¡young ¡wom·an of ¡hum·ble !dress (Sketch Book)
[48] Moran rose and ¡went out·!side. (Amongst Women)
Longer SENTENCES with longer WORDS can accordingly present an elaborate mosaic of stressed and unstressed SYLLABLES, as in [49].
[49] !Am·nes·ty ¡In·ter!na·tion·al is ¡in·de!pend·ent of all !gov·ern·ments, po·¡lit·i·cal !fac·tions, ¡i·de·!ol·o·gies, ¡ec·o·!nom·ic !in·ter·ests, and re·¡lig·ious !creeds. (Amnesty)
VI.16 The principal unit of organisation for PROSODY, as I noted at the outset (VI.1), is not the CLAUSE, as in the LEXICOGRAMMAR, but the TONE GROUP [Note 8] spoken as an integrated sequence, usually assigning one STRONG STRESS near the end (e.g. for END WEIGHT), and being set off by perceptible pausing in the PACE, however slight, before and after it. The UTTERANCE is spoken as an integrated sequence that constitutes a contribution to a DISCOURSE, and the approximate spoken counterpart of the written SENTENCE (cf. VI.1). Just as a SENTENCE may consist of a single CLAUSE, an UTTERANCE may consist of a single TONE GROUP, as in [50]; or of several, as in [51]. Here too, I signal a PAUSE (in the sense of VI.11) with an upright line | for a shorter one and two | | for a more decisive one, again relying on my own performances.
[50] “¡You re·!fuse?” | | “Of ¡course I re·!fuse.” | | “I ¡think you are ex·¡treme·ly !fool·ish.” | | “¡In·!deed!” (Piccadilly Jim)
[51] ¡Well, | ¡here’s a ¡boy that’s been a ¡reg·u·lar !fel·low — | ¡raised in A·!mer·i·ca — | done ¡work on a !news·¡pap·er — | ¡sud·den·ly ¡tak·en off to !Eng·land (same)
Making a TONE GROUP coincide with a CLAUSE, as in [52], is a handy strategy to match up PROSODY with LEXICOGRAMMAR. But coinciding with a NON-CLAUSE, as in the response in [53], can serve strategic functions too, such as lending WEIGHT to emotions.
[52] To·¡day’s my ¡on·ly !chance. || ¡Aunt ¡Car·o·line has ¡gone a·!way. | | ¡Fath·er will be ¡bus·y in the !gar·den (Damsel)
[53] “Your forget that I am the receiver of confessions.” “¡Oh !my! || ¡Yes, | !high ¡church. || !Why, | in the ¡name of !God!” (Wingless Bird)
VI.17 A whole TONE GROUP or UTTERANCE manifests a PROSODIC CONTOUR. Its most characteristic feature in English is the fluent rhythm arising from the distribution of STRESSES separated by unstressed SYLLABLES whose PACE is accelerated as they become more numerous — usually one, two, or three, as in [54-55].[Note 9]
[54] !Bless·ed is ¡he that !read·eth, | and ¡they that ¡hear the !words of this !proph·e·cy, | and ¡keep those ¡things which are !writ·ten there·!in: | for the !time is at !hand. (Revelation 1:3)
[55] It is a !truth ¡u·ni·!ver·sal·ly ac·!knowl·edged, | that a !sin·gle ¡man in pos·¡ses·sion of a ¡good !for·tune ¡must be in ¡want of a !wife. (Pride)
If all SYLLABLES were paced the same, the effect would lack BALANCE.
VI.18 The dominant PITCH CONTOURS in English are either falling from the first STRESS in the TONE GROUP down to the END [56]; or rising steadily from the FRONT up to the END [57].[Note 10] A falling PITCH CONTOUR can carry along the falling CERTAIN STRESS, whereas a rising CONTOUR can carry along the rising UNCERTAIN STRESS (VI.13). In return, a steady CONTOUR requires fitting the falling or rising STRESSES to the PITCH level at various places. So in a falling CONTOUR, an earlier STRONG STRESS will get higher PITCH than a later one [55], whilst in a rising CONTOUR, just the reverse occurs [56]. These STRESSES, displayed here for clarity as short arrows just above the long CONTOUR arrows, can remain distinctive with a louder VOLUME and a slower PACE.
VI.19 Graphic displays of PROSODY in fine detail are hindered by a trade-off: the more precise and elaborate the display, the more unwieldy it is to perform or interpret, and the more it can be influenced by individual differences in spoken performances, speech habits, language varieties, etc. Just displaying all four parameters of PITCH, STRESS, VOLUME, and PACE needs a graphic like [58] for Strider’s threat to the hobbits at Bree in The Fellowship of the Ring. I am assuming three TONE GROUPS with a short rising PITCH on the opening UNSTRESSED SYLLABLES and then a longer falling PITCH starting from the first point of STRESS and arriving at a STRONG STRESS for END WEIGHT. The first two TONE GROUPS move toward louder VOLUME and slower PACE; the final one is quite slow and unites falling PITCH with increasing VOLUME.
This is more staged than the way Viggo Mortensen says it in cinematic masterpiece of the vociferous Peter Jackson, but then his Aragorn -- except in battle -- is mostly a low-key, soft-spoken lord.