|
Department of English Introduction to English Language Online Unit 1: Phonology |
Browser Alert - Important Setting Instructions: Windows users will need Microsoft Internet Explorer version 4.5 or later (eg 5 or 5.5) to correctly display the International Phonetic Association symbols in these notes. Macintosh users will need Internet Explorer version 5 or later and may have to adjust the 'Character Set' setting, available under the 'View' menu - the 'Western (Mac)' set appears to work the best. If you don't have the right browser settings some or all of the symbols will appear only as small rectangles. If the following appear as characters rather than as rectangles, you probably have the right settings: T, I«
The purpose of this unit is to raise conscious awareness of the relationship between spelling and sound.
Consider the nature of rhyme. In poetry, rhyme is seldom across a whole word. The words love and above are said to rhyme because the last part of each word which has the same sounds. Sometimes whole words do rhyme. In this case, the spelling could be the same spelling repeated, e.g. plays (verb) with plays (noun) or different spellings, e.g. prays with praise (this is actually very rare - see further below). Two different words which have exactly the same sound are known as homophones.
Find as many differently spelled homophones as you can for the following words. You may like to type them into the space provided after each word and then check your suggestions against those which occurred to us. Click on the underlined word (the link) on the left to reveal our comments/feedback.
Homophones are words which have the same sound but different meanings and usually different spellings. This definition is slightly complicated by the difficulty of knowing when we are dealing with a "different word". An example of a homophonous pair with the the same spelling is bow/bow /bAu/, pronounced to rhyme with how /hAu/ (these symbols are explained later in this Unit). Let us say that bow 1 is an inclination of the body or head, and bow 2 is the fore-end of a ship or boat. It so happens that these two words have quite different historical origins, but that is less important than the fact that we think of them as being two quite different words.
Now consider another homophonous pair, this time with different spellings: bow/bough. Both bow 1 and bow 2 are homophonous with bough. In this case, it so happens that bow 2 is historically related to bough, but hardly anybody knows this, and even if they did, it would not stop them from thinking of them as two quite different words. A repetition of exactly the same word e.g. two occurrences of the word elephant, would not be regarded as an example of homophony. There are a lot of borderline cases.
Homophones are relatively uncommon in English compared to some other languages, and it is extremely rare to find homophonous words at the ends of two rhyming lines of verse:
The hooly blisful martir for to seke,
That hem hath holpen whan that they were seeke
[Chaucer, Canterbury Tales, General Prologue ll.17-18]
Swinburne rhymes gilt with guilt and bough with bow in A Dialogue. You may know other examples. One authority excludes this possibility from the notion of rhyme: 'Rhyme in verse is the repetition, in the lines of a poem, of the same end sound or sounds.....The repetition of sound must not extend forward to the initial consonant sound for then the effect would be one of total repetition or identity rather than rhyme.' (Stillman1966:32). This is a strict definition, which not everyone would agree with.
Rhyme as we understand it in English was not used in classical verse, and is unknown in the poetry of a number of other languages, e.g. Polish. Other kinds of sound repetition were, and are, widely used in the verse of most languages, including English. The technical terms describing different types of sound repetition (alliteration, assonance, pararhyme etc.) are explained in various books including Crystal 1987:74 and 1995:414. For a more detailed discussion of sound in poetry see Leech (1969:89-102)
How many vowels are there in English? At school we learn that there are five, but this only refers to letters of the alphabet, not to sounds. One conundrum much loved by children and some teachers goes: "Can you think of a six-letter word which has no vowel in it?" The answer, often delivered smugly, is rhythm. The obvious problem with this conundrum from a phonological point of view is that it does have a vowel. This happens to be represented by the letter y, but it is still a vowel. In the word spry, there is also a vowel sound, different from the sound in rhythm, but still represented by the letter y.
There are in fact about twelve vowel sounds in English, and some of these combine to make diphthongs (glides from one vowel to another), so that altogether there are about twenty vowel distinctions which speakers of English recognise. We say 'about' so as to take account of different accents of English. The letter y in the two words just mentioned represents the vowels /I/ and /ai/.
Consider the letter a. It is quite versatile, being used to represent a large number of different vowel sounds, depending on the word it is in and the letters it combines with, e.g. rat, rate, rather etc. Note down as many different vowel sounds involving the letter a as you can think of. For each sound, give a short list of words containing that sound. (You can use the space provided below to type in your answer, if you like.)
For example, thinking of the sound represented by a in rat, we can list words such as catch and bag etc.
So,
The twenty-six letters of the alphabet have to cope with about 45 sounds. This is what chiefly accounts for the alleged great irregularity of English spelling. However, the English spelling system does represent sound systematically, albeit in a rather complex way. This complexity is the result of historical change in English over hundreds of years and is certainly not something which can be easily simplified now that the process of historical evolution in spelling has been brought to a virtual halt by standardisation.
The idea that spelling is unsystematic is strongly influenced by well-known examples of two types of non-correspondence between sound and symbol.
First, many sounds can be represented by more than one letter of the alphabet, e.g. the sound at the beginning of the words site and cite.
This phenomenon is not confined to individual sounds, but also affects combinations of sounds. For example, if we add a third identically pronounced word, sight, we now have one combination of sounds represented by three different combinations of letters (see Activity 1). This is actually less troublesome than it appears. Young readers soon discover that combinations such as i + consonant + e (e.g. site), i + gh (e.g. sight) and -ied (e.g. spied) at the end of a word account for a very large proportion of words containing this sound, so it is not as difficult as it looks. It is quite likely that the phenomenon known as dyslexia includes a deficiency in the ability to make these systematic correspondences.
There are of course some further problems. Readers also have to apply their knowledge of stress patterning to learn the distinction between the two pronunciations of -ied in words such as spied and hurried. There are also exceptional cases where the i + consonant + e pattern represents a different sound (e.g. engine). These must be fitted in around the statistically frequent patterns, but most readers do this relatively easily. That is not to underestimate the difficulties faced by some people in learning to read, but merely to point out that spelling is not completely anarchic.
Second, many letters in the alphabet have more than one way of representing sound. We looked at the example of y in rhythm and spry above. Activity 2 dealt with the letter a. The letter i is just as complex. Think how it is pronounced in the following words: bid, bide, bird, boiled, bailed, nation, debris.
Moreover, just as a single combination of sounds may be represented by several different spellings, so a single combination of letters may represent several different sounds. Probably the best-known example in English is the combination -ough. There are arguably seven different pronunciations available, but in practice the distribution of words across these pronunciations is very uneven.
The situation can be summarised thus: (The symbols in slanting brackets / / represent sounds - see below.):
|
sound eg. ./s/
|
letter eg. "s"
|
||||
|
letter
|
letter
|
letter
|
sound
|
sound
|
sound
|
|
cite
|
site
|
kiss
|
site
|
rising
|
fusion
|
|
/s/
|
/z/
|
/S/
|
|||
The mismatch between spelling and sound has resulted in many calls for spelling reform, including ideas for expanded alphabets to take account of all the sounds in the language. For various reasons, and despite the enthusiasm of reformers, it seems improbable that reform will happen. Crystal 1987:216 provides a good summary of the issues.
Linguists need to be able to refer to sounds unambiguously, and thus need an agreed set of transcription symbols. The set used in this course is based on Cruttenden (1994). You are advised to stick to these, although you will come across other symbols, or the same ones used in slightly different ways, in other books.
Transcription symbols are similar to letters of the alphabet in that they are conventional signs representing the sounds of the language. Because English has many more sounds than the 26 letters can cope with, there are obviously more symbols than there are letters. However, a good number of the symbols are in fact identical to letters. This is chiefly because the people who devised the symbols were speakers of European languages, using the Roman alphabet. Remember that in becoming literate we have to master more than 26 alphabetic forms. For a start there are upper and lower case forms. Then there are handwritten and typewritten forms. This all gets sorted out in due course, but the early stages are slow. Transcription is similar; it is a task requiring accuracy, attention to detail and considerable practice. To transcribe in print, one needs special software. (To connect with a Web site from which you can download special font software which will enable you to use the IPA in word processors such as Microsoft Word, click on the following link Summer Institute of Linguistics, free IPA fonts
As we have indicated above English spelling is often an unreliable indicator of pronunciation. Sometimes there will be a direct correspondence between spelling and IPA symbol(s) and sometimes there will not. When learning the IPA and the principles of transcription it can be useful, therefore, to be aware of where these correspondences apply, and where they do not.
| IPA transcription | keywords | spellings | voicing / other features |
| /p/ | pie, apple, snap | p, pp | unvoiced |
| /b/ | buy, cabbie, tap | b, bb | voiced |
| /t/ | tie, cattle, cat | t, tt | unvoiced |
| /d/ | die, paddle | d, dd | voiced |
| /k/ |
kite (but also) tackle, sack (and also) cat |
k ck c |
unvoiced |
| /g/ | guy, haggle, hag (but also) guide, plague |
g, gg gu |
voiced |
| /f/ | fight, stifle, stiff (but also) photograph (and also) enough, cough |
f, ff ph gh |
unvoiced |
| /v/ | vie, survive, give | v | voiced |
| /s/ | sigh, hiss, castle (also many but not all plurals in s) hats, trucks, laughs (and also soft central 'c' as in) prince, dancer cancel |
s | unvoiced |
| /z/ | zoo, cozy, Oz (but also many plurals in s) hands, dogs, pans, fishes |
z | voiced |
| /h/ | high, ahoy | h | unvoiced |
| /m/ | my, gamble, hammer, Mum | m | voiced (nasal) |
| /n/ | nigh, banner, pan, candle | n | voiced (nasal) |
| /w/ | why, away | w | voiced |
| /l/ | lie, alley, pal, still, pale | l | voiced |
| /r/ | rye, arrow | r | voiced |
Note that there may be additional IPA to English spelling correspondances which I haven't listed here.
Double letters. Double letters (as in the two ps in apple, the two fs in stiff) typically only represent a single sound, so will only be transcribed by one IPA symbol: thus stiff (with two fs) = /stIf/ (with one /f/)
Problems with 'r' spellings. In many accents (for example, Received Pronunciation, Australian, and New Zealand English), word- final r and especially sentence/clause-final r is often not pronounced. Accordingly, if the following words were spoken carefully and in isolation, there would usually be no realisation of the final r spelling in the transcription: car, bear, hair, here, cure, her, etc. Thus car would be transcribed as /kA:/, with no /r/. Note, however, that in connected speech, final r will often be pronounced, even in so-call 'r'-less accents, when the following word begins with a vowel. Accordingly, the final rs in the following might well be pronounced and hence transcribed as /r/: 'your aunt', 'the car against the wall'
Problems with 'c' spellings. The IPA symbol /c/ is never used to transcribe an English sound. That is to say, you should never use /c/ when transcribing English. The letter 'c' typically represents either an /s/ sound, as in place (/pleIs/), prince and parcel, or /k/ as in cat (/kQt/) and cost. The /c/ sound occurs in other languages, but not in English.
Problems with the /z/ sound and 's' spellings. You should note that the /z/ sound is often spelt with an 's' letter. For example, the final sound in his is /z/ - /hIz/ The sounds /z/ and /s/ are both made with the tongue tip up against the hard palate behind the top teeth. They differ only in the feature of 'voicing'. The /z/ sound is voiced - the voice box is fully activated in its pronunciation. The /s/ sound is unvoiced - the voice box is not activated in its pronunciation, or, more accurately, it only comes into play relatively late in the pronunciation. You can hear the difference by alternately pronouncing his (voiced final sound = /z/) and hiss (unvoiced final sound = /s/). Accordingly, you often need to be wary not to let an 's' spelling influence your transcription when the sound is actually /z/. For example, the plural 's' following a voiced consonant is typically pronounced /z/ - for example hotels (/h«UtElz/, beds (/bEdz/).
Problems with the sound /j/ and 'y' spellings. The sound heard at the beginning of words such as yell, yacht and you is transcribed /j/ - thus /ju:/. There is no /y/ sound in English. You will see below that the spelling 'j', as in jacket, represents a sound which is transcribed as /dZ/
/T/ and /D/
/S/ and /Z/
/tS/ and /dZ/
/N/
Consonant sounds all involve either a complete or partial blockage of the airstream. Vowels, in contrast, all allow for a relatively free flow of air.
There are two different types of vowels in English called monophthongs and diphthongs. A monophthong is a vowel sound which has a constant quality, while a diphthong changes its quality - its starting quality is different from its ending quality.
The monophthongs of Received Pronunciation English (RP) are as follows
Short vowels
/I/ pit
/E/ pet
/Q/ pat
// pot
/Ã/ putt
/U/ put
/«/ patter
Long vowels
/i:/ bean
/Î:/ burn
/A:/ barn
/:/ born
/u:/ boon
Diphthongs
/aI/ bite
/EI/ bait
/I/ boy
/«U/ roe
/aU/ house
/U«/ cruel
/I«/ ear, beer, clear
/E«/ air, bear, chair
Cruttenden, A. 1994. Gimson's Pronunciation of English.. Edward Arnold.
Crystal, D. 1987. The Cambridge Encyclopaedia of Language. CUP.
Crystal, D. 1995. The Cambridge Encyclopaedia of the English Language. CUP.
Leech, G.N. 1969. A Linguistic guide to English Poetry. Longman.
Stillman, F. 1966. The poet's manual and rhyming dictionary. Thames and Hudson.
Crystal, D. 1995. The Cambridge Encyclopaedia of the English Language. CUP.
Chapter 17.
Kuiper, K. and W. S. Allan, 1996. An Introduction to English Language. Sound, word and sentence. pp. 19-27.
Roach, P. 1983. English Phonetics and Phonology. CUP.