The English Language: A Guided Tour of the Language

by David Crystal

  aI∫ʊbIspraIzdIfdƷɒnəmɛərIwəleIt

  This much more readily conveys the speed at which the sentence would be said, though it is much more difficult to read.

  However, if you compare this transcription with the one above, you’ll notice certain differences. The /d/ of should has disappeared. The vowel sound of be has changed. The first vowel of surprised has disappeared. The pronunciation of and has totally altered. And the vowel in were has been replaced. What is going on?

  These differences aren’t random. Changes in pronunciation are inevitable when we start to speed up our speech, and run words together. As we begin to speak a word, we use our vocal organs to make the first sound, but already our brain is planning how to make the second sound, and the third. This planning may be so advanced, in fact, that the brain may already have sent signals to the vocal organs telling them to get ready for these later sounds.

  So, in the above sentence, the lips are preparing for the /b/ sound of be long before the tongue has finished pronouncing the /d/ sound at the end of should. The faster we speak, the less time there is for the tongue to do anything at all. And there comes a certain point when we simply do

  A spectrograph is a machine which turns sound waves into a pattern of marks on paper. The above pattern is what emerged when I recorded the sentence ‘This is a Penguin book’ on a spectrograph. The vowels and consonants can be identified, as shown by the symbols. But note that the gaps between the symbols aren’t reflected by a similar pattern of gaps on the paper. The sound waves run continuously from one sound to the next, showing that the vocal organs are continuously in use.

  not try to keep the pronunciation separate. We allow our /d/ to come out as a /b/, or we just drop it altogether. In phonetics textbooks, when one sound is influenced by another in this way, it is called assimilation. When the sound is dropped completely, it is called elision.

  There’s another example of assimilation and elision later in the same sentence. When and is used before an /m/ sound, at normal conversational speed, the /d/ is dropped. (And is often spelled ’n or n’ in writing that tries to reflect the nature of informal speech.) The /n/ sound is also affected by the following /m/, and blends with it, to produce a single / m/.

  You can practise the effects of assimilation and elision very easily. Try saying the above sentence slowly, with every sound produced clearly and distinctly. Now say it faster, but trying to keep all the sounds in. Now

  faster still… There will come a point when it proves impossible to ‘get your tongue round all the sounds’. Something has to give. You will automatically start to assimilate and elide.

  Or again, you can show the way the brain anticipates later sounds by watching what happens to your lips when you say certain words in front of a mirror. Take the two words she and shoe. First, practise the sounds separately. Say the sh consonant by itself as if you were telling someone to be quiet. Note how you hold your lips. Then say the vowel of each word, still looking at the lips: first the /i:/ of she, then the /u:/ of shoe. Notice how the lips become very rounded for /u:/. Now say the whole word shoe, and watch what happens to the lips as you start to say the sh. They will start to become rounded in shape right away. This doesn’t happen when you say the whole word she. Say the two one after the other: she – shoe – she – shoe. One sh is lip-rounded, the other is not.

  It should be clear what is happening. The brain ‘knows’ that a vowel is due later in the word, and that it has to be pronounced with the lips rounded, so it sends instructions to the lips to get ready, right at the beginning of the word. As a result the boundary between the two sounds is extremely difficult to draw.

  This kind of thing is happening all the time in speech. The most dramatic effects take place when we speak very quickly, but even in fairly slow speech it is impossible to eliminate them. Try saying shoe as slowly as you can, and you will still notice the ‘pull’ on the lips as the /u:/ sound approaches.

  Here are some other sequences where the last sound of the first word is affected by the first sound of the second:

  that boy the /t/ sound changes into a /p/

  that girl the /t/ sound changes into a /k/

  this shop the /s/ sound changes into a /∫/

  In such sequences as next day, last chance, left turn, and kept quiet, the /t/ at the end of the first word is dropped. We actually pronounce lef’ turn, and it needs quite an effort of will to say the sequence with the first /t/ in. The same thing happens at the end of a word like cyclists. The t here is hardly ever pronounced in conversational speech.

  Sometimes, the two sounds influence each other, so that they both change. This happens in such sequences as would you, where the /d/ of would and the /j/ of you combine to produce /dƷ/ (as in the first sound of jaw). Similarly, in what you (as in I don’t know what you see in her), the two consonants combine to produce /t∫/ (as in chew). Informal speech What you doing? is often written so as to show this blend, using wotcha or watcher.

  And lastly, an example of a sound being added. In Received Pronunciation, the r spelling at the end of a word is not sounded when the word is said on its own: four, mother, care. But when it is followed by a vowel, at normal speed, the /r/ is pronounced: four o’clock. This is called a ‘linking r’. By making this link, speakers find it easier to pronounce the words in a smooth sequence.

  Pronunciation worries

  It is of course possible, on special occasions, and with special training, to speak English in a way which avoids making most of the effects noted above. Actors declaiming the poetic lines of Shakespeare would generally avoid them (though they would be present in the prose passages representing everyday speech). And generally actors try to pronounce words clearly, paying special attention to their endings, and without rushing weak syllables. In this way, the words can be heard more clearly at the back of the theatre. But their speech is much slower than normal conversation as a result. In three renditions of Hamlet’s ‘To be, or not to be’ speech by different actors, the speeds ranged from 130 to 190 spm – on average, half the speed of normal conversation.

  The same point applies to other professional voice users, such as news broadcasters, radio announcers, priests, judges, and politicians. These people all have to speak in abnormal conditions – from a pulpit, in front of a crowd, into a microphone. In most cases, the listeners cannot see the speaker’s face clearly enough (or see it at all, in the case of radio) to enable them to get clues from the movement of the lips, or from the facial expression, as to what is being said. To be sure that their speech stands the best chance of being understood by all, then, professional speakers know that they must speak relatively slowly and distinctly. If they do not, they risk criticism for being unintelligible, or for being too informal, casual, or ‘sloppy’.

  Enter the radio listeners, for the most common of all complaints to the BBC concerns the topic of pronunciation. And sloppy speech is the charge most often cited. The irony, of course, is that in almost every case the words called sloppy are in fact perfectly normal pronunciations in everyday speech, and everyone uses them. They include such forms as Feb’ry for February, lib’ry for library, Antar’tic for Antarctic, as’matic for asthmatic, twel’ths for twelfths, patien’s for patients, reco’nize for recognize, and so on. It’s very difficult in fact to say some of these words in their ‘full’ form – try pronouncing the second t in patients, for example. But many listeners, it seems, expect such precise articulation over the air, and are ready to demand it in writing, to the tune of thousands of letters each year.

  Most listeners give just one reason for their complaint: a letter is there in the spelling, and so it should be pronounced. This is another example of the widespread belief, mentioned above, that speech is a poor relation of writing. We always need to remind ourselves that speech came first, in the history of our species, and that we all learn to speak before we learn to write. To be worried about our pronunciation because it does not match the spelling is a strange reversal of pri
orities. We also need to remember that pronunciation patterns have changed radically since the days when the spelling system was laid down. English spelling hasn’t been a good guide to pronunciation for hundreds of years (see Chapter 5).

  But despite all this, many people do get very angry when sounds are left out that they think ought to be there, or sounds are put in which they think ought not to be. Probably the most famous case of this last point is the use of an ‘intrusive r’ by speakers of Received Pronunciation: the insertion of an /r/ between vowels, when there is no r in the spelling. The most well-known instance, because of its frequency in the news, is law and order – widely known as ‘Laura Norder’.

  One listener sent in a collection of over 100 intrusive rs which he had heard in one day’s listening. He included examples like Shah(r) of Persia, draw(r) ing, and awe(r)-inspiring. These are the noticeable ones, because the /r/ stands out clearly after the ah/aw vowels, which are said with the mouth quite widely open. It’s much more difficult to hear this kind of /r/ when it occurs after the less sonorous /ə/ vowel – the vowel that we use at the end of words like sofa or Persia. Unless Received Pronunciation speakers are taking extreme care, and speaking very self-consciously, they automatically put an /r/ into such phrases as Africa(r) and Asia, an area(r) of disagreement, and drama(r) and music. I have a tape recording of a critic vociferously condemning the intrusive r in law and order, in the course of which he said ‘the idea of an intrusive r is obnoxious’, putting in an /r/ at the end of idea!

  Where does the intrusive r come from? It’s the result of these speakers unconsciously extending a pattern already present in their accent, as found in the linking r sequence described above. It is important to notice that, although there are thousands of English words which end in the letter r, only four kinds of vowel are involved: /ɔː/, as in four, /aː/ as in car, /3ː/ as in fur, and /ə/ as in mother. What has happened is that, over the years, the linking /r/ has been extended to all words ending in one of these four vowels, when they’re followed by another vowel. The effect is most noticeable in words ending in /ɔː/, as in law and order, because there are in fact not very many such words in the language, so the usage tends to stand out.

  Of course, explaining why a pronunciation has developed doesn’t explain why some people have come to hate it. It’s the same with other areas of usage. Why do some people hate hopefully (see p. 31) ? The reason is likely to be something to do with the way one social group, at some time in the past, adopted a usage in order to keep themselves apart from another social group which did not. In particular, an accent comes to be used like a badge, showing a person’s social identity. At any one time, there are several pronunciation patterns which are ‘loaded’ in this way. Longstanding examples include ‘dropping the h’ (’ospital for hospital) and ‘dropping the g’ (walkin’ for walking). These days such forms are considered to be uneducated – though a century ago, g-dropping was often to be found in cultured speech (as in the upper-class use of huntin’, shootin’ and fishin’).

  All of this presents radio managers with a problem, of course. Although only a minority of listeners are antagonized by such matters, none the less they are antagonized – and this is not what radio broadcasting is meant to be about. Announcers and presenters who are sensitive to these issues therefore often go out of their way to avoid using a pronunciation which they know will upset some people. They may go through their scripts and underline problem cases. Far more than the intrusive r is involved, of course. The problems include changes in stress (e.g. dispute vs dispute) and the pronunciation of individual words (e.g. saying recognize with or without the g). A selection of issues is given on p. 62. Foreign words pose special problems, as do the names of people and places. A Pronunciation Unit has long been established at the BBC to help answer

  Controversies of recent years

  The following list includes many of the words which have alternative pronunciations in current English. The asterisk indicates the pronunciation recommended in the 1981 BBC guide compiled by Robert Burchfield.

  adversary stress on *1st or 2nd syllable

  apartheid vowel in 3rd syllable as in height or *hate

  apparatus vowel in 3rd syllable as in car or *fate

  applicable stress on *1st or 2nd syllable

  ate vowel as in *set or late

  centenary vowel in 2nd sllable as in ten or *teen

  centrifugal stress on *2nd or 3rd syllable

  comparable stress on *1st or 2nd syllable

  contribute stress on 1st or *2nd syllable

  controversy stress on *1st or 2nd syllable

  deity vowel in 1st syllable as in say or *see

  derisive s in 2nd syllable as in *rice or rise

  dilemma vowel in 1st syllable as in *did or die

  diphtheria ph as *f/ or /p/

  dispute stress on 1st or *2nd syllable

  economic vowel in 1st syllable as in met or me (both accepted)

  envelope vowel in 1st syllable as in *den or don

  furore said as *3 syllables or 2

  homosexual vowel in 1st syllable as in *hot or home

  inherent vowel in 2nd syllable as in *see or set

  kilometre stress on *1st or 2nd syllable

  longitude ng as in *range or long

  medicine said as *2 syllables or 3

  migraine vowel in 1st syllable as in *me or my

  pejorative stress on 1st or *2nd syllable

  plastic vowel in 1st syllable as in *cat or car

  primarily stress on *1st or 2nd syllable

  privacy vowel in 1st syllable as in *sit or sigh

  sheikh vowel as in see or *say

  Soviet vowel in 1st syllable as in *so or cot

  status vowel in 1st sllable as in sat or *state

  subsidence vowel in 2nd syllable as in Sid or *side

  trait final t *silent or sounded

  queries about such matters. During the 1986 World Cup matches, the Unit had to issue guidelines to its commentators about the players, referees, linesmen, managers, and others involved – it took twenty-seven pages.

  The problem becomes particularly acute if a radio channel decides to adopt a policy of friendliness or informality in response to listener demand. To make speech come across in a normally informal way, it is necessary to speed it up, and to introduce assimilations and elisions. If these are not introduced, either because they lessen the clarity of what is said or because they attract listener criticism, the speech will inevitably sound formal, clipped, and controlled. But listeners cannot have it both ways. If they want their announcers to sound friendly, they must expect a chatty style, with all the consequences this has for pronunciation.

  As long as society contains divisions, there will always be differences in pronunciation, and, as a consequence, arguments about which form is best and which accent is most acceptable. The arguments can be healthy and informative, or nasty and intolerant. They are usually the latter. BBC announcers with accents other than Received Pronunciation have been known to receive hate mail. And when I presented English Now on Radio 4 in the 1980s, my own accent – a mixture of Wales, Liverpool, and southern England (see p. 88) – was often criticized. The letter-writers usually ask for the removal, forthwith, of the offending parties.

  Getting the sack because of your speech isn’t unknown. I know of two cases – one in an estate agent’s, the other in a hairdresser’s – where assistants have had to leave because their accents were felt to be inappropriate. And in 1970 there was a much-publicized case of a blacksmith who committed suicide because he could not cope with the ridicule levelled at his accent when he moved from Yorkshire to the South of England. Remembering such stories, a tiny plea for tolerance would seem a reasonable way to end this section.

  Received Pronunciation

  In England, there is one accent that has come to stand out above all others, traditionally conveying associations of respectable social standing and a good education. This ‘prestige’ accent is k
nown as Received Pronunciation, or RP. It is often associated with the south-east of England, where most RP-speakers live or work, but in fact it can be found anywhere in the country. Accents usually tell us where in the country a person is from; RP tells us only about a speaker’s social or educational background.

  The ancestral form of RP developed in the late Middle Ages, in London and the south-east, as the accent of the court and the upper classes. It was well established over 400 years ago. The Elizabethan courtier George Puttenham, writing in 1589, thought that the English of ‘northern men, whether they be noblemen or gentlemen… is not so courtly or so current as our Southern English is’. Some courtiers did hold on to their local speech – Walter Raleigh kept his Devonshire accent, for instance. But most people anxious for social advancement would move to London and adopt the accent they found there. As a result, the accent soon came to symbolize a person’s high position in society.

  During the nineteenth century, RP became the accent of the public schools, such as Eton, Harrow, and Winchester, and was soon the main sign that a speaker had received a good education. It spread rapidly throughout the Civil Service of the British Empire and the armed forces, and became the voice of authority and power.

  Because RP had few regional overtones and was more widely understood than any regional accent, it came to be adopted by the BBC when radio broadcasting began in the 1920s. During the Second World War, the accent became fixed in many people’s minds as the voice of freedom. The terms RP and BBC English became synonymous.

  These days, with the breakdown of rigid divisions between social classes and the development of the mass media, RP is no longer the preserve of a social élite. It is now best described as an ‘educated’ accent – or perhaps ‘accents’ would be more precise, for there are now (and maybe always have been) several varieties. The most widely used variety is that generally heard on the BBC. But in addition there are both old-fashioned and trend-setting forms of RP. The trend-setting variety is often described as ‘far back’, or ‘frightfully, frightfully’ – the ‘Sloane Ranger’ accent of the 1980s. The more conservative variety is found mainly in older speakers – what is sometimes referred to as a rather ‘plummy’ tone of voice. You’ll hear it in the recordings of BBC plays or announcements from the 1920s and 1930s.