Global versus Analytical stresses

The motives for accenting a word or syllable in English include the following:

(1a) Highlighting of contrasts generally overrides all other tendencies to assign deliberate, intended, voluntary (‘accentual’ ) stress to a word or syllable.
We may define an accent (or accentual stress) as one consciously and voluntarily accorded by a speaker to a particular word or syllable. It shd be noted that a pitch movement which is the latter element of a complex tone,  if so delayed that it is effected even on a subsequent word, since not having being chosen to specifically emphasise that word, does not constitute an accent.

(1b) The perhaps uniquely strong English-language tendency to deny even a syllable within a word (not just a word) its normal stressing in favour of another syllable not normally stressed but conveying or emphasising a contrast is in general never revealed in the written language. However, an extremely rare example of doing so was to be seen in the Journals of Arnold Bennett (1954 p. 88) in which we find “I saw few signs ... of suppressed or expressed excitement...” where the writer conveys it by italicisation of a prefix. 

Other examples include: Maˈjorities and ˋminorities. Midday is normally /mɪd`deɪ/ but midnight /ˋmɪdnaɪt/ yet either pattern may be reversed in contrastive contexts.

A typical example of a usage in which a NNS (non-native speaker of English) may fail to observe the custom of highlighting semantic contrast occurs with the use of a beginning like In my `country... when a contrast is intended with another person’s country. In such a situation a native English-speaker would normally only say In `my ˏcountry... Similarly NNSs offen fail to notice that the stressing of A friend of mine, in spite of the suggestion of contrastiveness that mine seems to embody, is usually idiomatically A `friend of mine and is rarely found stressed A friend of `mine unless extreme direct contrastiveness is involved as with Yes he `is a friend of `ˏmine | but I `didnt know he was a friend of `ˏyours `ˏtoo.

(2a) Avoidance of Re-accenting of Re-occurrences. A most powerful application of the ‘rule’  of highlighting of contrasts is seen operating inversely when speakers deny stress to a word etc which reappears quite soon in a discourse. This avoidance of re-accenting of re-occurrences ie of words or even merely syllables which are identical or constitute or embody exactly the same reference is a generally obeyed rule for native English speakers. An example of a ordinarily unaccentable syllable in fact usually being stressed would be in speaking of the subject of a journal as being ˈAlcohol and Alcohol`ism.

(2b) A syllable generally considered to be completely unstressable by most speakers may very occasionally be heard accented eg It was an exˈciting and excita`ble /ɪksaɪtə`bəl/ performance. Another related accentual possibility for English native speakers reverses this pattern of retrospectively avoiding re-accenting. This occurs as anticipatory avoidance of accenting the usual syllable in a word in order to highlight an accentual contrast with a subsequent word of a phrase. Another example is the beginning of a sentence thus: One of the `irritations, but at the same time `fascinations, of the traditional orthography of English... where the ordinary accentuation of each of the words irritation and fascination would be on a later element as irri`tation and fasci`nation but, in order to csubmit to the strong inclination to highlight the contrast between the two words’ initial syllables, a speaker may sometimes depart from their normal accentuations.

In an exchange like Wd you like ˈtea or `coffee?" the reply  I’d like `coffee, ˏplease’ of course re-accents the word ‘coffee’ but the preference to avoid re-accenting of re-occurrences is overridden by the need to accent ‘coffee’ for the sake of contrasting ‘coffee’ with ‘tea’.

(2c) Whole phrases may sometimes be given no accent if they contain immediately repeated matter. For example a speaker who has said ‘They’ll have nowhere to hide may receive the response ‘Yes in`deed they’ll have nowhere to hide or ‘Of `course they’ll have nowhere to hidewith the repeated they’ll have nowhere to hide constituting only an intonational tail (usually at a low level pitch) ie being completely without an accent. This is another example of where it would be possible to let the final word rise but that final movement wd be perceived by native-English speakers not as an independent accentual Rise but as the delayed completion of a Fall-Rise complex tone (and so completely non-accentual). The tonetic notation Yes in`deed they’ll have nowhere to ˏhide wd make the pitch movement quite explicit but a tonological notation of the type of expression that we exemplified above (such as appeared in OConnor-&-Arnold 1973) wd show a sentence with exactly that set of pitches as Yes in`ˏdeed they’ll have nowhere to ˳hide making the non-accentual value explicit by using the symbol [˳] to indicate where the rise is to be understood to take place. In these notes tonetic transcriptions are normally supplied, tonological ones being resorted to only on occasions where a special need seems to exist such as the preference in some cases not to insert marks of complex tone completion within the body of a word, eg preferring oc`ˏcasionally to oc`casionalˏly or oc`casioˏnally.

Its also possible for a second speaker to respond making a long (low) prehead of the repeated words thus They’ll have nowhere to hide in`deed. This procedure avoids having any accents in the repeated phrase by uttering the whole of it at the same low level pitch.

(2d) In the same category as avoidance of re-accentuation of words referring to or representing ideas, facts etc already referred to, mentioned or ‘given’ is the preference not to accent words conveying ideas, facts or circumstances already well known to both speakers including matters, conditions etc of which both are fully aware such as the state of the weather or the situation  they are in.

(3) Rhythmic preference

If an expression doesnt involve avoidance of re-accenting of re-occurrences, the default tendency for speakers is to stress  last suitable (ie more semanticly charged, content-conveying not merely grammatically functional) words or syllables in any intonational phrase or sentence. These will most often be  nouns.

Speakers may prefer to avoid intermediate stresses either depending on how rapid they want to make an utterance or in compliance with the rhythmic preference for stress alternation as in 'Let me have a bit of something to `eat or 'Let me 'have a bit of 'something to `eat in which (i) all potential stresses between the first and last words are suppressed and (ii) the potential stress on bit is suppressed: if it were employed it would produce a heavily insistent suggestion. See also Division 4 of this website where in  Section 7 at Honorifics ¶139 other examples are given of ALTERNATE STRESS PREFERENCE. See also in the same Section 7 Prepositions at ¶84.

4) Globalisation. The preference for a latest or earliest-and-latest stressing pattern one may call  or globalising, or simply global stress distribution.

(5a) Analytical stress distribution

At the phrase level, English-speakers mostly  prefer analytical stress distribution with its regular suppression of stresses on anaphoric expressions such as initial articles like this and final enclitic pronouns such as it.

An example of English speakers departing from a more analytical approach in order to treat globally a very common expression in which stress on the contrastive word own would be usually expected is 'Mind your own `business. Contrast the analytically accented synonymous 'Pay attention to your `own affairs and items like That's `my business. This example is remarkable because the more logical alternative stressing *Mind your `own business can be said to be very  unusual . Similar expressions like It’s none of our `business can however,  alternate with the more predictable stressing especially if it’s softened by use of a falling-rising tone viz Its `none of `ˏour  ̥business.

(5b) In the rather special context of an ‘insistent rising head before a Fall climax tone the usual stressing practice may be counteracted eg (i) ˏMind  your ˏown ˋbusiness. (ii) It says ˏsmoking or ˏnon ˋsmoking.

Many expressions may occur in either the globalising or the analytical form eg the ˈbest of ˈboth `worlds or the ˈbest of `both worlds.  The preference in such cases may be due not to the avoidance of re-stressing a word (or synonym for a word) but simply that an idea is present in the consciousness of the speaker. This can occur for example when the first speaker has seemed to have treated a subject with inadequate seriousness. Though no actual joke has been made the response may just as possibly be Its no joking `matter as the  stressing Its `no `ˏjoking   ̥matter. The same type of explanation applies to expressions like There's ˈno doubt a`bout it which might be said when the speaker expects disbelief on the part of the collocutor.

 It must be remembered that these are only  tendencies which in practice may be forsaken by the individual speaker for a variety of reasons some of which it is impossible to discover eg notably when they reflect what is going on in the speakers mind but is undeclared. Even more notably, any word representing any idea, fact or circumstance of which both speakers in a conversation are fully aware may very often be denied an accent. Examples of this might be the name of the place where theyre conversing, the state of the weather or the political situation.  Departures from such usual practices amount to perhaps at least ten percent of utterances for most speakers of English. Anyone who listens out for such departures will be almost certain to hear a number of them during the course of a day spent listening to a variety of conversations, unscripted broadcast speech and the like. Its a natural but unfortunate consequence of the effects of pedagogical accounts of the patterns of English intonation that teachers are sometimes inclined to criticise departures from their ‘rules rather too harshly.

 Some kinds of expression may show the tendency to highlight contrast not strong enough to prevail over the tendency to globalisation. This happens with eg the saying to get the wrong end of the `stick. This might well be expected to climax on wrong or end but its climactic final accent is usually placed on stick. Similarly one might have expected the usual stressing to *the boots on the `other foot but in fact its normally globalised to the boots on the other `foot.

8) Another of the reasons for the departures may be that the speaker has simply been subject to what, in regard to non-prosodic features of speech, would simply be termed a slip of the tongue. A peculiarity of speakers’ treatment of such prosodic ‘mistakes is that, unlike the way people treat mistakes of articulation, which are very often corrected by the speaker, it is only very rarely indeed that one hears a repetition that amends the prosody of an utterance just prosodically mis-managed. After all, theres no recognised way in writing of correcting any aspect of prosody other than word stressing by underlining, italicising or capitalising the word in question. Any purely pitch-pattern errors can hardly be represented in normal writing at all. Its noteworthy also that people are completely untroubled by having various prosodic features removed or even replaced by something linguistically quite inappropriate when they hear the words of a song, tho the best and most satisfying settings are those that most closely accord with the speech rhythms that would be naturally used in merely saying the words.

9) Another factor one must bear in mind is that by temperament one individual may be far more heedless than another about prosodies in ordinary situations. In circumstances where the speakers wish to speed up or slow down or are influenced by distracting contexts or by circumstances such as alcohol consumption the tendency to depart from their normal prosodic practice is increased — sometimes even by at least partly losing the thread of what they wish to say.

10) Animation Stresses. Certain stresses people use from time to time may seem to be intended as inappropriate ways of highlighting individual words when that is not the purpose for which the speaker has adopted the stressing but rather as a device for increasing the intensity of a whole expression (phrase, sentence etc). I long ago proposed for such usages the term ANIMATION STRESSES. At the word level, anomalous stresses that might be termed ‘prosodic slang’ are used by many speakers when eg bra`vo, ra`ther, yip`pee etc are so stressed and by some children (at public schools at least at one time) saying the Latin warning shout  ca`ve /keɪ`vi/. The American usage posi`tively may well be generally viewed as or at least have begun as an accentual extravagance. That was possibly the perception of it by people who first herd the adverb which is canonically `absolutely but has now become in emphatic use perfectly normal as `abso`lutely. It’s possible for an empty’ word like thing or matter to be accented merely to amplify the force of a sentence eg in Its ˏnot a ˏgood `thing. A strong indication that the speaker is not truly accenting a word in the normal sense of accentuation occurs when a word like it which can reasonably be said to never be accented in a non-contrastive context is given a major stress as in `Its all ˏright. The reader who turns to Section 4.1 on this website can hear the actress (free to choose her own prosodies) in Item 15 line 3 say `Its a 'pale `blue [Cf our Blog 032 of 17 June 07.]

(11a) One should remember that persons reading aloud or acting, and thus using not their own spontaneous choices but prescribed wording, are very prone to prosodic mistakes – the more so the less their performance has been prepared. Ive been shocked on various occasions to observe that the director of a play or film has allowed a performer to employ a completely inappropriate prosody. I can’t remember any occasion when anyone has ever indicated to me that they received such a shock, tho the basis on which a dramatic performance is considered good or not must often be in large part influenced by the aptness or otherwise of the actors’ prosodic choices. For the very rare phenomenon of a writer referring to such a matter we have to turn again to the Journals of Arnold Bennett (1954 p. 215 ed. F. Swinnerton) where he mentioned of a Troupe of about 40 that Not one could avoid the most elementary false emphasis. Thus Sylvia May looking at a man asleep on sofa, 'But he may wake up' (when there was no question of another man asleep) instead of “'He may wake up. Arnold Bennett was a very successful playwright and director of his own plays. Among very numerous examples one might give of actors inappropriate stressings there is, from the Orson Welles film of Othello, She might ˈlie by an ˈemperor’s ˎside (instead of the more effective She might ˈlie by an `emperor’s side. In the film Ghandi the doubtfully suitably stressed even if you `caused a good deal of trouble was followed immediately by the grossly inappropriate accentuation "Especially if you `caused a great deal of trouble" where spontaneous speech would be totally unlikely not to have an accent on deal.

(11b) Certain kinds of inappropriate accentuations are very commonly heard from radio and tv newsreaders whose scripts often contain synonymous references back to such things as the names of cities at which the reader fails to de-accent the synonym. For example A ˈpowerful ˋbomb was set off at ˋX--- this morning | and a ˈhuge ˈcloud of ˎsmoke ˈstill ˈhangs ˈover the ˋcity. Here at city there is plainly no reason for a suggestion of any contrast with another area so it should not be accented or at least it should be denied a full Fall. A weaker climax may occur with only a low fall which may sound less grossly inappropriate but it still doesnt sound completely normal. They may even re-accent a word on its repetition from being obliged to by a badly written script. Ive found that something like “The ˈwater ˈlevels | have reached ˈdangerous ˎlevels” to be far from uniquely clumsy. That type of expression can occur in both spontaneous and scripted speech.

(12a) Its possible for a word to be accented twice in a sentence because of semantic re-focusing eg because a contrast of meaning is involved when the word re-occurs in a different sense. This has been called ‘anta`naclasis’ (OED 'A figure of speech .. when the same word is repeated in a different, if not in a contrary signification'.

(i) If you `want to make good `ˏcoffee,| you must use plenty of `coffee.

(The first “coffee is the beverage, the second the beans or powder.)

(ii) There are `ˏpalaces | and ´`palaces. (ie ordinary ones & specially fine ones.)

(iii) Its not what she ˋˏsaid | it was the way she ˊˋsaid it.

The first said refers to the semantic content of the words used, the second to to the manner of their delivery (“It's not `what she ˏsaid...” would have been a more explicitly meaningful way of expressing the idea).

(iv) Boys 'will be ˋˈboys. (ie young men inevitably exhibit the behaviour of immature males)

(v) 'Tolerance | is what 'makes 'Britain ˋBritain. (ie makes the country the kind of country it is)  Tony Blair (December 2006)

(vi) Making the unˋˏmissable|unˋmissable. BBC slogan advertising their postponed-listening facility by which a broadcast that has not been able to be received at its original transmission time may be received by computer at a time chosen by the user usually during the following week. (2008)

(vii) Robert Burns referred to ˈman's inhuˈmanity toˋman (people's cruelty to their fellow human beings)

(viii) See also People Speaking 4.1.2 line 5 where 'Who was that lady  I saw you with last night?' receives the reply 'That was no lady. That was my wife' This very old joke turns on semantic re-focusing. The sense of lady intended by the first speaker is merely 'woman', but the sense it’s supposed to be taken as is 'woman of refinement'.

(ix)  ˈDog eat ˋdog is used to describe competition that is as extreme as cannibalism. The first dog is elliptical for "a dog"; the second is the customary reference to an animal etc as foodstuff without any article.

(x) In Kingsley Amis's Lucky Jim the eponym says that his boss is "the lousiest Professor of History in history". ('History' ie the subject; 'history' ie the period of time.)

(xi) It's the regular usage with items of botanical and zoological nomenclature to give the generic name first and the specific second accenting them both even when the two require the same word. Compare ˈBellis pe`rennis, ˈVulpes, `vulpes” orGoˈrilla, go`rilla”.

(xii) One presumes that German speakers as well as English speakers may well say that someone " `isnt just plain `Mr ˏStein | but `Dr `Dr Stein".

(12b) Note also: An ˈeye for an ˏeye |and a ˈtooth for a ˋtooth. A ˈbargain is a ˎbargain (well-known proverbs etc). 

Kinds of re-focusing etc can perhaps explain the following also: Let ˈbygones be `bygones. The ˈblind leading the `blind. People said it wouldn't sur`vive | but sur`vive it ˋ ˏhas.  If the ˈworst (ˈ)comes to the ˋ ˏworst...  ˈBusiness is ˋbusiness. Let's share and share a`like. `Amy,| being `ˏAmy | wouldn't  a`gree. The ˋonly thing we have to ˏ fear | is ˈfear itˋself. Let's call a spade a spade.

There are cases where people seem to be rather vague or inattentive about operating the "rule" as can be seen in some common sayings. `place for ˏeverything | and ˈeverything in itsˋplace is usually so accented. This can be but usually isn't accented A `place for ˏeverything | and ˈeverythingˋin its place.

It's also possible for words to be accented on immediate re-use in certain other situations such as when a speaker “echoes” another’s words nominally at least for confirmation, as in the following.

I want some `money. – `You’re asking `me for ´money? (You must be `mad. `I’m a `pauper. `ˏYou,| are `rich.)

(13) (i) Another type of re-accentuation within the same phrase may occur when a word is repeated immediately for emphasis eg: (a) big big job, dear dear (me), (a) long long time, (a) lovely lovely day, many many times, (the) old old story, a red red rose, never never say that,  really really beautiful, very very nice, etc.

(ii) Where a word is separated from its repetition by only a particle, the word is regularly accented on both occurrences. Examples include: again and again, for ever and ever, an eye for an eye, arm in arm, back to back, blow by blow, day by day, face to face, from ear to ear, from strength to strength, heart of hearts, home from home, hope against hope, inch by inch, more and more, neck and neck, night after night, on and on, from time to time, wheels within wheels (also possibly ˈwheels with`in wheels). Note also expressions like She ˈkeeps her(ˈ)self to her`self .

In Parliament on 12 June 1990 MP Gerald Kaufman said "The Government ... is isolated on |ˈissue| after  ˈissue | after ˎissue."

(14) Although these "analytical" sentence stressing tendencies observable in English-speakers are very strong and may in some cases make a listener uncomfortable about what was a speaker’s precise meaning or whether a speaker has succeeded in expressing their meaning properly etc, in regard to these “rules”, which as we say, can be heard to be broken every day, EFL teachers need not concern themselves unduly if they find themselves baffled in their natural desire to understand what brought about such infringements. On the other hand they should not neglect to inform their more advanced students of the spoken language of the existence of these overwhelmingly often firmly followed patterns of behaviour to be found in all the principle varieties of spoken English employed by educated users.

(15) Finally, there are plenty of expressions in English that contain accentuations which can hardly be explained on logical principles – what we are obliged to term as accentuation idioms. Among the many examples one could quote are the following: It’s none of your `business, That’s all there is `to it, Think nothing`of it etc. These are only in a limited sense idioms because they can usually be attributed to the speaker’s reacting not to a verbal formulation by the interlocutor but to something that the speaker perceives as indicated or adumbrated by the other person. The word doubt may not have been used by a speaker but what has been said may be taken as expressing dou't in our example above. As we've said, a speaker may avoid accentuation of the word laugh not because that word has occurred in the exchange but because the interlocutor has laughed or even smiled (when we have exaggeration by the speaker) eg in the sentence It’s ˈno laughing `matter. There are some other cases for which it's difficult to perceive a logical explanation. For example, despite the usual powerful feeling the speaker has that a contrast must be highlighted, it is still idiomatic to stress eyes in the back of one's `head with no stress on back. As far as the EFL user is concerned, it's advisable not to worry about such items but to take them to be idioms. Getting them "wrong" is in any case usually of very little consequence.

Despite our insistence that it's very possible for re-accenting of re-occurrences to sound very strange to the ears of native English speaking people, almost any day one may come across what can only be described as completely unaccountable breaches of the 'rule' — in fact probably as often as at least in ten percent of situations where one would expect it to be complied with. For   example, it shouldnt be thought that it was bad acting or lax direction that could explain the fact that the very successful actor James Stewart in the 1958 Hitchcock film Vertigo answers his girl friend who says Come on! `Tell. with There's nothing to `tell. It seems hardly possible to refer such apparently 'unnatural' linguistic behaviour to inattention.