3 Describing English consonants

3.1 Whats inside a phonetic symbol?

So far, we have considered the IPA essentially as an alternative writing system, which allows us to express a larger range of sounds than the English spelling system would. However, looking only at those sym ols might suggest that we are dealing with individual, self!contained units when we consider phonemes and allophones" each is like a locked lack ox la elled with an IPA sym ol. In fact, each IPA sym ol is shorthand for a whole range of properties, and those properties explain how the particular segment eing sym oli#ed is pronounced$ unpacking the lack ox for each sound reveals not a %um le, ut an internal structure, and understanding that structure allows us to make comparisons with other sounds. &hen we know that'k(, for instance, is a voiceless velar plosive, we can start to see what properties it shares with other sounds which might also e voiceless, or velar, or plosives$ we can also see how it differs from other sounds which are not voiceless, or velar, or plosives. )urthermore, we shall see what properties different allophones of the same phoneme share, which might allow them to e regarded as *the same+ y speakers of English" that is, we can work out what particular phonetic features speakers of English tend to ignore, and which they are aware of. Since this may e very different for speakers of other languages, unpacking IPA notation in this way also allows cross!linguistic comparisons to e made. In this chapter, we shall therefore consider a very asic set of phonetic features which ena le us to descri e the articulation of the consonants of English, and to assess their differences and similarities.
3.2 Consonant classification

A iologist looking at some particular creature wants to know various things a out it, to work out where it should e placed in conventional iological classification. Some properties are visi le and therefore easy to work out, such as how many legs it has or whether it has fur, feathers or scales. In other cases, closer o servation will e needed" tooth shape cannot usually e checked from a distance. Still other properties are ehavioural, and our iologist might need to o serve her creature over a longer period of time to figure out whether it lays eggs or ears live young, or what it eats. ,he same goes for phonetic classification" some properties are straightforwardly o serva le when you look in a mirror, or can e figured out easily from feeling what your articulators are doing. -ther features are harder to spot, and need some extra training efore you will ecome aware of them. )urthermore, we also need to remem er that phonemes are realised as various different allophones, so we must uild up a picture of all the possi le environments where that phoneme can occur and what happens there, to sort out how it ehaves. .iologists today are, of course, working within an agreed classification" when they o serve a creature with particular physical traits, or particular ehaviours, they can slot it into a framework of her ivores and carnivores$ mammals, insects, irds and reptiles$ verte rates and inverte rates$ and so on. )ortunately, phoneticians and phonologists have a similar, generally agreed framework for sounds. )or consonants, we need to know six things to arrive at a classification" in the rest of this chapter, we shall consider these six sets of properties in turn, and assess which English phonemes fit into each category. /owel classification involves rather different features, and we return to this in 0hapter 1" we are eginning with consonants ecause many of their properties are easier to ascertain from self!o servation, and ecause the systems of consonant phonemes in different accents of English vary far less than the vowels.
3.3 The anatomy of a consonant 3.3.1 What is the airstream mechanism?

Speech is audi le ecause the movements of articulators 2to e discussed in su se3uent sections4 cause the air to vi rate, forming sound waves which travel to the hearer+s ears, and set up vi rations in her inner ear, which are then translated into sounds again y the rain. Since sound waves need air, it follows that articulatory vi rations will only make sound waves if there is a moving ody of air availa le. Airstreams can e set in motion, or initiated, in three ways$ however, only one is used in English, and indeed is found in every language of the world. Essentially, speaking is modified reathing" it makes use of the resources involved in normal respiration, ut in a more controlled way. &hen we are simply reathing 3uietly, the phases of reathing in and out last approximately the same time, and expiration is not under our physicalcontrol$ it simply occurs as an automatic conse3uence of having reathed in. However, when we are speaking, the phase of reathing out is significantly longer, depending on the length of the utterance we want to produce. A network of muscles, like the intercostal muscles etween our ri s, come into play to make reathing out smoother, more gradual and more controlled during speech, providing a regular flow of air which can then e modified y the articulators in various ways. All the sounds of English, oth consonants and vowels, are produced on this pulmonic egressive airstream, where the initiator is the lungs and the rest of the respiratory system, and the direction of airflow is outwards" this is overwhelmingly the most common airstream mechanism in every language of the world. It can generally e taken for granted that the sounds under discussion elow are pulmonic egressive, ut you should remem er to give that information in a complete description" so the la ial nasal 'm( 2which, as we shall see, is produced using the lips 5 hence la ial, and with airflow through the nose 5 hence nasal4, is strictly a pulmonic egressive la ial nasal. It is possi le to produce speech using a pulmonic ingressive airstream. 6o language seems to use this airstream regularly for particular sounds, although it has een reported in various cultures as a means of voice disguise" if you try to reathe in and speak at the same time, you will find that the pitch of your voice raises significantly. ,here are two other airstreams which may e involved in speech, although even in languages where these are used, they will characteri#e only a few sounds, interpolated in a stream of pulmonic egressive speech. ,he first is the glottalic airstream mechanism, initiated y a movement of the larynx, which is where you can feel your *Adam+s apple+ protruding slightly a out half!way up your throat. ,he larynx can move up or down, and the glottalic airstream can therefore e either ingressive or egressive, producing sounds known as implosives and e%ectives respectively$ none of these occur in English. )inally, the *tut!tut+ click sound ' 7( is produced on a velaric airstream, which operates only ingressively. &hen you make ' 7( you can feel that the ack of your tongue is pressed against the roof of your mouth, stopping air from moving any further ack$ a little air is then drawn into the mouth further forward, and the closure with the tongue is released to make a click. 6either the glottalic nor the velaric airstreams provide airflow with the volume or controlla ility of the pulmonic system.
3.3.2 Voiced or voiceless?

A ma%or division among speech sounds which is relevant for all languages is the dichotomy of voiced and voiceless. If you put your fingers on your *Adam+s apple+ or *voice ox+ 2technically the larynx4, and produce a very long '#######(, you should feel vi ration$ this shows that '#( is a voiced sound. -n the other hand, if you make a very long 'sssssss(, you will not feel the same sort of activity" 's( is a voiceless sound. Pulmonic egressive air flows through the trachea, or windpipe, and up into the larynx, which is like a mo ile little ox suspended at the top of the trachea, acting to control the airway to and from the lungs, with the epiglottis a ove it protecting the lungs y stopping foreign odies like food from dropping in. Stretched across the larynx from front to ack are the vocal folds, or vocal cords. ,hese can e pulled ack and drawn apart, in which case they leave a free space, the glottis, through which air can flow" this is the case for voiceless sounds like 's(. )or voiced sounds, the vocal folds are drawn together, closing off the glottis$ however, the pressure of air flowing from the lungs will cause the folds to part, and their

essentially elastic nature will then force them together again. 8epetitions of this cycle of opening and closing cause vi ration, as for '#(. ,he num er of cycles of opening and closing per second will depend on the si#e of the vocal folds, and determines the pitch of the voice" hence, children+s smaller, shorter vocal folds produce their higher voices. Although sounds can e voiced in any position in the word, voicing is most o vious medially, etween other voiced sounds" when there is an ad%acent voiceless sound or pause, voicing will not last for so long or e so strong. 0onse3uently, although English has the minimal pairs tip 5 dip, latter 5 ladder, bit 5 bid for 9t9 versus 9d9, 'd( is only voiced throughout its production in ladder, where it is medial and surrounded y voiced vowels. &ord!initially, we are more likely to identify 9t9 in tip y its aspiration, and 9d9 in dip y lack of aspiration, than rely on voicing. /oicelessness and voicing are the two main settings of phonation, or states of the glottis" for English at least, the only other relevant case, and again one which is used paralinguistically, is whisper. In whisper phonation, the vocal folds are close together ut not closed$ the reduced si#e of the glottis allows air to pass, ut with some tur ulence which is heard as the characteristic hiss of whisper.

3.3.3 Oral or nasal?

,he next ma%or issue is where the pulmonic egressive airstream used in English goes. )or most sounds, air passes from the lungs, up through a long tu e composed of the trachea, or windpipe$ the larynx$ and the pharynx, which opens out into the ack of the oral cavity. ,he air passes the various articulators in the mouth, and exits at the lips$ and all these vocal organs are shown in )igure :.;. However, for three English sounds, air passes through the nasal cavity instead. ,he key to whether air can flow through the nose is the velum, or soft palate, which you can identify y curling the tip of your tongue up and running it ack along the roof of your mouth until you feel the hard, ony palate giving way to something s3uashier. )or oral sounds, the velum is raised and pushed against the ack wall of the pharynx, cutting off access to the nose. However, for 'm(, 'n( and ' <( in ram, ran and rang, the velum is lowered, so that air moving up from the lungs must flow through the nose. If you produce a long 's(, you will e a le to feel that air is passing only through your mouth$ conversely, if you hum a long 'm(, you will notice that air continues to flow through your nose while your lips are pressed together, with that closure eing released only at the end of the 'm(. &hen someone suffering from a cold tells you *I+ve got a cold id y dose+ instead of *I+ve got a cold in my nose+, she is failing to produce 'n( and 'm( ecause soft tissue swelling locks air access to the nose and perforce makes all sounds temporarily oral. 6asal sounds, like 'm( and 'n(, are produced with air only passing through the nasal cavity for at least part of their production. -n the other hand, nasalised sounds, like the vowel in can, preceding a nasal consonant, as opposed to the vowel in cat, which precedes an oral one, are characterised y airflow through oth nose and mouth simultaneously.
3.3.4 What is the manner of articulation?

,o produce any consonant, an active articulator, usually located somewhere along the ase of the vocal tract, moves towards a passive articulator, somewhere along the top. &here those articulators are, determines the consonant+s place of articulation, as we shall see in the next section. How close the active and passive articulators get, determines the manner of articulation. ,here are three main manners of articulation, and one su sidiary case which in a sense is intermediate etween the first two.
A. STOPS

If the active and passive articulators actually touch, stopping airflow through the oral cavity completely for a rief period, the sound articulated is a stop. If you put your lips together to produce 'p( pea, and hold them in that position, you will feel the uild!up of air which is then released when you move from the stop to the following vowel. )urther ack in the vocal tract, 't( tea and 'k( key are also stop sounds. =ore accurately, all these are plosives,

the term for oral stops produced on a pulmonic egressive airstream, %ust as clicks are stops produced on a velaric ingressive airstream, for instance. Plosives may e voiceless, like 'p(, 't( and 'k(, or voiced, like their e3uivalents ' (, 'd( and 'g(.Since the definition of a stop involves the complete, transient o struction of the oral cavity, it also includes nasal sounds, where airflow continues through the nose. English 'm(, 'n( and ' <( are therefore nasal stops, although they are typically referred to simply as nasals, as there are no distinctive English nasals involving other manners of articulation. All these nasals are also voiced. )inally, some varieties of English also have su types of stops known as taps or trills. &hile a plosive is characterised y a complete o struction of oral airflow, followed generally y release of that airflow, a tap is a very 3uick, allistic movement where the active articulator strikes a glancing low against the passive one$ interruption of the airstream is real, ut extremely rief. =any Scots speakers have a tapped allophone '( of the phoneme 9r9 etween vowels, as in arrow, very$ many American speakers have a similar tap as a realisation of 9t9 in butter, water. ,rills are repeated taps, where the active articulator vi rates against the passive one. ,rilled 'r( is now rather uncommon for speakers of English, although attempts at imitating Scots often involve furious rolling of 'r(s.
B. !"CAT"#$S

>uring the production of a fricative, the active and passive articulators are rought close together, ut not near enough to totally lock the oral cavity. ,his close approximation of the articulators means the air coming from the lungs has to s3uee#e through a narrow gap at high speed, creating tur ulence, or local audi le friction, which is heard as hissing for a voiceless fricative, and u##ing for a voiced one. English 'f( five and 's( size are voiceless fricatives, while 'v( five and '#( size are voiced. ,he su class of affricates consists of sounds which start as stops and end up as fricatives$ ut as we shall see in 0hapter ?, they ehave as single, complex sounds rather than se3uences. Stops generally involve 3uick release of their complete articulatory closure$ ut if this release is slow, or delayed, the articulators will pass through a stage of close approximation appropriate for a fricative. ,he two relevant sounds for English are //, at the eginning and end of church, and its voiced e3uivalent//, found at the eginning and end of judge. If you pronounce these words extremely slowly, you should e a le to identify the stop and fricative phases.
C. APP!O%"&A'TS

It is relatively easy to recognise a stop or fricative, and to diagnose the articulators involved, since these are either touching or so close that their location can e felt. In approximants, on the other hand, the active and passive articulator never ecome sufficiently close to create audi le friction. Instead, the open approximation of the articulators alters the shape of the oral cavity, and leads to the production of a particular sound 3uality. ,here are four approximant consonant phonemes in English" 9%9 yes, 9w9 wet, 9r9 red 2although as we have seen, 9r9 may have a tapped allophone for some speakers4 and 9l9 let. All these approximants are voiced.
3.3.5 Is the airflow central or lateral?

,his parameter is rather a minor one, since it distinguishes only one phoneme of English from all others. )or almost all English consonants, the airflow through the oral cavity is central. 8ecall that fricatives, like 's( or 'f (, are produced with close approximation of the active and passive articulators$ however, if you produce any fricative, you will feel that your articulators are actually pushed together 3uite tightly at the sides of the oral cavity, with the actual close approximation, and hence the narrow gap for airflow, left in the middle. ,he same is true for all the approximants except one" if you produce rip and lip, and focus on the initial consonants, you will notice that while the outgoing air for 9r9, as usual, moves along the centre of the mouth, for 9l9 it moves down the sides. If you find this difficult to feel, try making the related voiceless fricative sound found in &elsh names spelled with

@llA, like Llewellyn$ ecause this is a fricative and involves close approximation of the articulators, the airflow is easier to o serve. Alternatively, try making an 'l( ingressively, pulling the air into your mouth instead of reathing it out, and feel the cold air moving inwards along the sides of your tongue. In English, oth the clear and the dark allophones of 9l9, and only these, have lateral airflow, and are known as lateral approximants. Since the only case where the central versus lateral difference is distinctive in English involves 9r9 and 9l9, these should consistently e descri ed as central and lateral respectively. Although in a particularly thorough description, all other sounds 2except nasals, which have no oral airflow at all4 should e explicitly stated to e central, this definition will generally e understood rather than stated elow, since the other English sounds do not contrast with lateral sounds of the same place and manner of articulation, meaning that confusion is highly unlikely.
3.3.6 What is the place of articulation?

As we have seen, the location of the active and passive articulators determines the place of articulation for a consonant. In English, consonants are produced at eight places of articulation. Since we have now covered all the other articulatory parameters re3uired to descri e consonants, introducing and defining these places will allow us to uild up a complete consonant phoneme system for English. In the ta les elow, the phoneme or allophone in 3uestion is initial in the example word, unless another part of that word is old!face.
A. B"(AB"A(

)or a ila ial sound, the active articulator is the ottom lip, and the passive articulator is the top lip. 9p9 pie voiceless ila ial plosive 9 9 by voiced ila ial plosive 9m9 my voiced ila ial nasal ,here is at least one further English phoneme which to an extent fits under this heading" this is the approximant 9w9 in wet. In producing 'w(, the lips are certainly approximated, though not enough to cause friction or o struct the airflow$ ut you should e a le to feel that the ack of your tongue is also unched up. ,his additional articulation takes place at the velum, so that 'w( is not simply a la ial sound, ut a la ial!velar one. 9w9 witch voiced la ial!velar approximant
B. (AB"O)*$'TA(

)or la io!dental sounds, the active articulator is again the ottom lip, ut this time it moves up to the top front teeth. 6ote that these sounds are la io!dental, while 9w9 and 9 79 are la ial!velar, ecause in the first case, articulation takes place only at a single location, while in the second, there are two separate, simultaneous articulations. 9f9 fat voiceless la io!dental fricative 9v9 vat voiced la io!dental fricative
C. *$'TA(

In most English sounds, and most speech sounds in general, the active articulator is part of the tongue$ to avoid confusion, places of articulation where the tongue is involved are therefore generally called after the passive articulator. )or the two dental fricatives, it follows that the passive articulator is the top front teeth$ the active articulator is the tip of the tongue. ,he tongue itself is conventionally divided into the tip 2the very front4$ the lade 2%ust ehind the lade, and lying opposite the alveolar ridge4$ the front 2%ust ehind

the lade, and lying opposite the hard palate4$ the ack 2 ehind the front, and lying opposite the velum4$ and the root 2right at the ase, lying opposite the wall of the pharynx4.

/ / thigh voiceless dental fricative // thy voiced dental fricative

*. A(#$O(A!

Alveolar sounds are produced y the tip or lade of the tongue moving up towards the alveolar ridge, the ony protrusion you can feel if you curl your tongue ack %ust ehind your top front teeth. 9t9 tie voiceless alveolar plosive 9d9 die voiced alveolar plosive 9n9 nigh voiced alveolar nasal 9s9 sip voiceless alveolar fricative 9#9 zip voiced alveolar fricative 9r9 rip voiced alveolar central approximant 9l9 lip voiced alveolar lateral approximant ,he sym ol 9r9 is used for the phoneme here and throughout the ook, primarily ecause it is typographically convenient$ ut different realisations of 9r9 are found throughout the English!speaking world, and as we have seen, 'r( itself, the voiced alveolar trill, is rather rare. ,he tapped realisation, '(, is also alveolar$ ut another even more common pronunciation is not. ,his is the voiced retroflex approximant, ' (, which is produced with the tip of the tongue curled ack slightly ehind the alveolar ridge$ this is the most common realisation of 9r9 for speakers of Southern Standard .ritish English and Beneral American.
$. POSTA(#$O(A!

If you move your tongue tip ack ehind the alveolar ridge, you will feel the hard palate, which then, moving further ack again, ecomes the soft palate, or velum. Postalveolar sounds are produced with the lade of the tongue as the active articulator, and the ad%oining parts of the alveolar ridge and the hard palate as the passive one. ,hey include two fricatives, and the affricates introduced in the last section. 99 ship voiceless postalveolar fricative 99 beige voiced postalveolar fricative 9t9 chunk voiceless postalveolar affricate 9d9 junk voiced postalveolar affricate
. PA(ATA(

Palatals are produced y the front of the tongue, which moves up towards the hard palate. &e have so far encountered two palatal sounds" the approximant 9%9 in yes, and the voiceless palatal stop 'c( in kitchen. 8ecall, however, that 'c( is the allophone of 9k9 found efore certain vowels$ velar 'k( appears elsewhere. ,here is a similar pattern for 9g9, which has as allophones velar ' ( in garden and palatal '7( give. Since we are constructing a phoneme system here, these allophones are not included in the list. 9%9 yes voiced palatal approximant
+. #$(A!

)or velar sounds, the active articulator is the ack of the tongue, and the passive articulator is the velum, or soft palate. ,he la ial!velar approximant 9w9 is not included here, as it was discussed a ove with the ila ials$ however, it should e remem ered that this dou ly! articulated sound strictly elongs under oth headings.

Similarly, although the *dark l+ realisation, 'l(, is also velar, it does not appear in the list elow as it is an allophone of 9l9. ,here is a further accent difference involving velar sounds" in some varieties of English, nota ly Scottish ones, there is a voiceless velar fricative, 9x9" this is the sound at the end of Scots loch, which speakers of other accents typically replace with a 'k(. 9k9 cot voiceless velar plosive 9g9 got voiced velar plosive 9<9 rang voiced velar nasal 9x9 loch voiceless velar fricative
,. +(OTTA(

Blottal sounds are in the minority in articulatory terms, since they do not involve the tongue" instead, the articulators are the vocal folds, which constitute a place of articulation as well as having a crucial role in voicing. English has two glottal sounds. ,he first is allophonic, namely the glottal stop, '(, which appears as an intervocalic realisation of 9t9 in many accents, as in butter. ,he glottal stop is technically voiceless, though in fact it could hardly e anything else, since when the vocal folds are pressed together to completely o struct the airstream, as must e the case for a stop sound, air cannot simultaneously e passing through to cause vi ration. ,he second, the voiceless glottal fricative 'h(, is a phoneme in its own right. 9h9 high voiceless glottal fricative

$-e.cises
;. 2a4 &hich of the following words egin with a voiceless fricativeC hang dogs cut ship chip foot zip sit 2 4 &hich of the following words egin with a voiced soundC nap jug knock lot pet jump fin 2c4 &hich of the following words ends with a stop soundC nap hang jug nudge bet lamb lots 2d4 &hich of the following words ends with an alveolar soundC pot sad boss lamb lamp size hen call 2e4 &hich of the following words contain an approximant consonantC wash hall map sing sigh red yellow D. 2a4 &hat do the initial consonants of these words have in commonC wash let right yet wish rough 2 4 &hat do the final consonants of these words have in commonC hop hot pass wish rough lock scratch 2c4 &hat do the initial consonants of these words have in commonC fish ship zip sigh house view :. ,ranscri e the words elow. ,hen write as full a description as you

can of all the consonants in each word, in your accent. )or instance, in doze 'd( is a pulmonic egressive central voiced alveolar stop$ '#( is a pulmonic egressive central voiced alveolar fricative. 8emem er to pay attention to the sounds, and not to the spelling. psalm jester which climb heavy splint loch bought squelch

!ecommendations fo. .eadin/

-f the text ooks recommended in the last chapter, >avenport and Hannahs 2;EEF4 provides the most accessi le and comprehensive introduction to articulatory phonetics, as well as a useful chapter on acoustic phonetics, which is not dealt with here. Some useful general introductions to phonetics are 8oach 2DGG;4, which may e of special help to non! native speakers$ .all and 8ahilly 2;EEE4$ 0atford 2;EFF4$ and Hadefoged 2;EF:4. ,he most comprehensive account of our current understanding of phonetics is Haver 2;EEI4. 8eferences relating particularly to the IPA were given in 0hapter ;.