I

-

'.

-

1

~

1

-

;1

;.....

rI

....

,1

t ""

~.

t

,,.;

J'

...

t

w

I

I '"

~.

I

DAN MATE-EBeU

ENGLISH PHONETICS

AND

PHONOLOGICAL THEORY

20th Century Approaches

~

Edi,tlllra Universitlit ii din Blllcure~ti

2003

Referenti otiintifici; Com. dr. :Ileana BaCiu <

Conf. dr. AndreiA. Avram

.:: ;

10 Erutum Universitllf;; din Bucur-e

$05. Panduri, 90-92, Bucure:lti -76235; TelefonIFax: 410.23.84

E-mail: editUra@unibuc.ro

Internet: www.editura.unibuc.ro

Tebnoredactare romputerizati.: Victoria lacob

Deserierea CW BibUJ)~i Nationale

I MATEESCU,DAN English phODotiO aDd pbODOlogiaJ an.lIT- 20tb century appnuu~bes I DM MateescuI BuCW'C~: F.di~ Univci.sifApi din Bucnre.

I 2002 208p. Bibliogr.

I ISDN 973;75-670-6 SlLlll'34I

J

] JJ 1

Heaven and earth shall pass away, but my words shall not pass

away. (Matthew, 24, 35) fI

PaloDins: What do you read, my lord? Hamlet: Word.'!, words, words (Hamlet, n, 2) [J The other (project) was a scheme for entirely abolishing all

words whatsoever: and this was urged as a great advantage in point of health a< well as brevity. For, it is plain, that every word we .

,

CONTENTS

;- Ack:lI.QwledgemeJ1is ............~......... , .... ".w w."..........u ...... 3" ...~......................w_

Chapter 1: IntroductloR................................................................................. .. 13

1.1. Language- 8. fundamental dirnenaion ofour existence ..........._ 13

1.2. The lost paradise of the original linguistic unit;y. The Babel

myth and subsequent nostalgia ..........,._......... , ...... _ ..... w_...... .. 15

1.3. Language as sound and meaning. The linguistic sign. Ferdi~

mmd de Saussure and Louis HjelmsJev ._.........................._ ... . 17

1.4. Language in the process of COl)llIllmication ....................... 24

1.5. Ll!I1gUage and writing ............................................................. . 27

Cbnpter 2: Artlcula/;)ry. AlIdit

$ I

~

i:'JE:=:z;;:::::::::::::=::::::::::::::::::::::::::::::::::::::::: 74 81'1 77 D. The F..ngJjsh Aifricates .............__...__.................. __............... . 85

t

~

Chapter 4: The Vowelll of Ehglal.. AN ArticNkrIury CJ_flCIdiol1.

Acoustic Correlates. Tl.. Pescription _ DislriJJJlIinn of

.. English MotlOpllthongs cmdPiphthongs .: .. :................................. . 89

4.L The Vowels: Critnria fur Clitssification .................................... 89

t 4.2. The Cardinal Vowel Cbru1s ..............__ ..... __............................... 94

-

4.3. EngJisb VoWels. The descciptioll and distribution of BIlgllsh

monophthl>ngs llJ1d diph1l:uJngs ................................................ 98

A- BIlg:l:ish simple 1'Owels ..,..................................................... 98

.. BIlgllsh li'ontllowels ....................................................... 98

h. BIlg:l:ish ba8

6.5. Palatalization ........................................................................... . 1496.6. L1liIlOl1S lH1d fortitions ............ , ............................................... . 1526.7. Delitions and insertions' ........................................................ _."' 1546.8. Metatlresis ................................................................................ .

157

Chapter 7: lJe}'IJ1Ulilie seglni!nI. SyUnbI. !drUeture in Ellglis/, ................... .

7.1. The Syllable; a furu:lamyllable. Phonotactie Cllll&trafuts .............. . 1627.3 The importance ofsegmental sonority for lbe ->yllable structure 1717.4. Coustraiots on onsets ............-.~.:.__ ...".h.......__._............_. ,.

1737.5. Constraints on coda.< ............................................................... .

1777.6. Syllabic consonants. Non-vocalic nuclei ................................. . 1827.-7. Syllabification in English ........................................................ . 182

Chapter 8: Supra.vegnreJJLa/ PIIOROInW: Stress, Rf,ytltm, lntonatwn ........ ..

8.!. Stess IlIld prominencc: The pbonemic (COlltn!Stive) function of 187

stress ....................................................................................... .

8.2. Free stress and fixe,d stress. The predictability of accentual 187

patterns ................................................................................... . 189

8.3. Metric paltems ......................... , .............................................. .. 190

8.4. MOIphological processes and stress ~bift ................................. . 1918.5. Primary and secondary strcss ................................................... . 1928.6. Weak and strong fonns. Vowolr~onanddeJitjon 1938.7. Rhythm .................................... ; .............................................. .. 1948.8. Intonational contoun;. ThcirP"'!:1lllrtic value ......................... . 195

l1lbliogT

~.

~,

~I

ACKNOWLEDGEMENTS J It has long been a custom to prefix to any product of our rrrind 1

a long (if not altogether endless) list of grateful thanks to the people ,~ who have assisted the author in bringing the respective work into the world. However long the list may be, the author is usually careful 1and cautiously adds that if someone was left out, this was only

because ofan unexplainable and unpardonable slip of the memory or

because, if exhauStive, the list would never end, indeed.

Consequently, excuses are made in advance to prevent reactions as I that of the evil (because disgruntled) fairy in the famous Sleeping

Beauty story. The list is also an opportunity of displaying the

numerous intellectual affinities that the authors have and a wonderful I occasion of introducing their family to the potential . reader,

undoubtedly ignorant of the author's luck of having such unique and

wonderful people around. IThough my intellectual and emotional debts are as great as any

person's, I would :rather not begin by thanking Plato and my great

grandparents for their contnlmtion in shaping my rrrind or my being. I

will mention only three persons to whom I feel immensely indebted I not only for their help during the various stages in the composition of this book, but also for the fact tha(they have always stood by me "in my most need" as Knowledge allegedly stands by Everyman. I am I grateful to professor Alexandra Comilescu for her constant affection, generous friendship and uninterrupted guidance; to professor ileana Baciu for her friendship, encouragement and advice; and last, but not I least, to Andrei A. AvraIn for being a living proof that selfless friendship, nobility of character and intellectual excellence can coexist in the same person. I

I

II

,

..

1

J

..

J

I

w

I

,J

1

~

1

-

1

-

-

~I

'1

1 ....

-

1

...

~I

:..

-I

-

CHAPTER 1

INTRObUcn:ON

1.1. Language':'- a fundamental dimension of our existence

The study of !an,,"Uage haS been 11 COlL~ preoccupation with more or less professional researchers for thousands ofyears. Since the earliest times, much before the birth of lingllitics as a distinct scholaIly diBcipline, people have been' aware of the essential role 1anguage plays not only in their evtnyday life, but also as a clJaracteristic :feature of mankind, radically differe.ntiating huwan beings from other species of the animal kingdom. The earliest religious or sacred texts record the importance given by our ancestors to language and their consciousness of the fuet that human existence itself cannot be conceived of outside the domain of language. If we were to refur only to the biblical tradition, tile, very beginning of the BOOK of the Genesis records the fact that the act of creation itself is intimately fuJked to speech. The C!e.!Ition oflighi is achieved through a speech act while all the subsequent stages of creation are preceded by God's fommJating His idea about what He was going to achieve, the creation proper only taking place, a.fu:r God pronounces the magic furtnula: "Let thcre be..."I The newly created realities need names and God, explicitly satisfied with His work, duly proceeds to the naming of His world. Creation is therefore preceded by, perl'onned through, and followed by, a linguistic expression, which thus anticipates, mate1ializ"s and completes it After the creation of man,

1 "And God said 'Let fuere be light' and the", W311Iighi" __ .And God said 'Let them be a firmamJmt' ...All1I God made the finnamenL __All1I God CI1I1ed the fi:rmameJJt Hoav""n (Genesis, I, 4. 6,7,8)

13

God created the other living. creatures and asked man to find names for thenL2 Thus Adam, who is roade "io the image of God" takes over from Him the important role of iostitutiog realities by means of language. The very act of creation cannot consequently be separated from that of naming whatever has come to have existence, sioce without a name the newly created realities don't achieve a full ontological status. The Gospel of Saint John notoriously begins by. equating the divioe creator of the universe with the "word", the Greek word logos (word) beiog actually a synonym for God (more precisely for one of the Persons io the Trinity, His son, Jesus ChriSt), for the source of whatever exists.' The essence of the myth seems thus to be .tha1 the whole world around us is language-based and thatlangoage is actually the very source of the existence of the wbole universe. A universe outside language, a universe wbere realities .don't have names, seems to be an utter impossibility.

The fact that lan"auage acts as a fundamental link between oUrselves and the world around us and that io the absence of language our relation to the Universe and to our fellows is dramatically impaired is something that people have been (a11east iotoitively) aware of since the beginning of history. Suffice it to mention that different coltures seem to associate speech problems with iotellectual deficiencies. See, for example,the meaniog of dunib (stupid) in English or the pretty similar situation of words like ''brubllit'' or ''fonf' io Romanian The origin oflan"ouage (believed to be divine iomost ancient coltures), the relation between language and thinking, the question if we can tbiok without the help of language (and ifwe can, what kind of thinking is that), the manner io which human beings (who are not, obviously, born with the ability to speak, but have, however, an ionate capacity

2 "And out of the ground the Lord Godfurrned every beast of the field, and every fowl of tim air; and brought them lIDto AdHm to see _ he would call them: and whatsoever Adam culled every living creature, that was the nam~ thereof'" (Genesis, 2, 19) .

3 "In the beginning was the Word, and the Word was with God, and the Word was God. The Word was in the beginning with God. All the things were made by him; and withoulhim WlISnat anything made that was made" (John, 1,1-3)

-,

J

for langUage acquisition) come, with an amAzing rapidity, to

SIlccessfully use language, beginning with the very first stages of their

existence (the aCquisition of language actually parallels the birth of the I child's self-conSciousness and the latter can hardly be imagioed

without the former) have puzzled researchers for centuries and none of

these questions has actoally received a satisfactory and universally I accepted answer.

I1.2. The lost paradise of the original linguistic unity. The Babel myth and subsequent nostalgia I

Scientists and students of language have been confronted,

sioee the earliest times, with an apparent paradox: on the one hand, the I diversity of the languages spoken all over the world (several

thousands are known) is really impressive, on the other hand, io spite

of the enormous differences among human idioms, they display I striking and fhndmnental similarities. To quote agaio the biblical

tradition, the famous legend of the Babel tower speaks about a unique,

origioallangoage, spoken by the first genetations ofhumans.(Genesis, I11). It was only God's jealousy and His fear that human beiogs, if

united, could really succeed io their attempt of building a tower that

should reach heaven, representing thus a real and very dangerous Ichallenge to divioe power that put an end to the golden age of the

lioguistic unity of mankind.4 No longer speaking the same idiom and

failing thus to understand one another and to SIlccessfully Ico=unicate among them, human beiogs ceased to be a significant threat to God. Theirlanguage is "confounded" and they are "seattered upon the face of all the earth". The unitary vision of the universe of Ithe golden age waS thus lost for ever and the uniiyiog and coherent

I4 "And the Lord said 'Behold, the people fu one and they have one language... and now nothing will be restrained from them, which they have imagined to do. Go to, let US go down, and there confuund their languages, that they may not understand one. anothera speecb'

)

J J J 1 J 1 .l

1

-

1

'"

...1

1

~.

~I

w

,,-,

image and understanding af!he world aroUlld us was replaced by a kaleidoscopic, multi-celoured one. 5 - _

Otb.e.mess has not always been a source of CQ!WCl:ll or of nostalgia fur the lost unity. It .has often been assumed to represent the mark of orunistakab1e identity, the basis of proud superiority in relation to !he oiliers. The ancient Greeb_considered their own language as ilie supreme manifesta.tion af human :intellect and the most appropriate tool :for undetstimding and coping with the universe. The other nations, speaking languages different from Greek were called "barbarian". Linguistic difference (and hypotOOtica1sUperiority) was thus the foundation of national pride, the ve.ty essence of what made their _ruct:i:on distinct from and better ethan the others. The etymology of the word itselfappartlIltly leads us to a root meaning "to stammer". Tlw foreigner was ilien someone unable to speak properly, his idiom was an impertect vehicle for human ilieught 6

The att~mpt, in ilie [lISt haIf of the 20th century, of creating-an arti:ficial language, universally spoken (esperamo) was jn fact a reflection of the ancient Babel myth, an expression -of ilie nostalgia for ilie lost onitaIy world of ilie beginnings. The divine ''cl1ISe~ seemed to be, however, still very powerful and initial eniliusiasm soon gave place to discouragement in frout of ilie obvious failure of ilie whole enterprise, confronted with enormous difficulties.

The Babel "syndrome" also manifested itself when, errthusiastic about the discoveries of comparative linguistic;; at the end

uof the 19 century and having collected irrefutable evideuce that the

!i Howe.ver~ as Umberto Eco remarks in a documented study deYoted to the "utopia ofa perfuct language", the diversi1;Y oflruxnan languages -llJld even races -had alreody been !lllIDtiou.e

are practicaJ1y inseparable). For Saussure, any linguilllic sign is made up of a signifumt iEn.,aUsh: signifier). tIu!t is an "acoulllic ima"ae" (the phonolol.,>ical "skeleton" of the word) and a signifie (English: signified), or a concept, to which the respective acoustic image sends.' Vile shoUld not mistake, however, Saussures's" image acoustiqne" for the real sourids we produce when we utter a .word. The Swiss linguist:himself warns against possible misinteI]lCCtations of his' theory. In spite of being mare concrete 1ban the cancept, the acoustic image is primarily a pSYChologic and not a material reality, which is proved, he argues, by the act tIu!t we C/lll speak to ow:selves without actUally articulating the words whose aconstic image is only present in our mind.'

Two are the essential feafures of the' linguistic sign in Sa:ussure's opinion: its arbitrariness and the linearity of the signifier. The "arbitrariness" of the linguistic sign has been one of the most fiunoll5 and heatedly debated of Sanssnre'ScQI1cepts. What he aebJally understands by the arbitr.n:iness of the sign is the arbitrariness of the relation holding between its constituent Parts. the signifier and the signified. 'This link is arbill:aty in .the sense that there is no reason whatsoever for which a particular string of sounds should be associated with a certain meaning. On the other hand, S!!JlSSUIe cautiously wm:ns against any mislll1derstanding of biB te:rminology. The associstion between the acoustic image and the concept is arbitrllr-y in the sense that it lacks motiva1imL; it is not arhitrmy, however, in the scMe t!wt it depends Dn the free. choice of the speakers. In reality, he argues, we have the veIy opposite sitnstion: once this association established, it becomes immutable; that is it cannot be changed. Langn.ages tend to be VeIy conservative systems and it is not up to 'any ofthe speakers in a linguistic commlllJity, and, indeed, not even to the eritire collectivity itself; to clIange the association between the sigtiifiers and the. sigcifif',ds in the Jan.,nuage

. Ii *Le sig;n.e linguistique unit non un.e rooSt: et un noIO, mats un concept et une image acoUJ1rique." (SalW;u:re: 1965 : 98)

9"(L'image RCaustique) n'cst pas Ie son lJllItenel, chose puremeut physique, mm. l'empreinte psycbique de ce son, la reprtsenmtion que nons en donne Ie temoignage de. nos sens.d (Sauasure: 1965 :98)

J

they use. For any linguilllic community language is somethihg

inherited, functioning on the basis of I"""" which the users of the

language cannot modify. Ibis doesn't mean, of course, 1bat languages J are fossilized systems, .given Once and for all. The changes they

undergo, however, take place over long periods of time and it is only,,

historical perspective thai enables U5 tn identify and analyze these I changes. Even so, no linguistic change C/lll. be given a "birth

certificate" where the eli:.act time of its coming into being and those

who "fafh.ered" it are mentioned.10 I

As fur as the arbitrariness of the linguistic sign is concemed,

two are, accordi.ng to the Swiss linguist, the situations in which We can

talk about some sort of match between the acoustic image and the I concept it is associated with: the onoma:topoeias and the exclamations.

Within the fIrst category, Saussure distinguishes between words th!lt

contain suggestive sounds ("des sonorites suggestives") which can I very well be the haphazaid result of phonetic changes, and "gennine"

onomatopoeia.~, in the ca~e ofwhich we can also speak only about an

approximate and partly conventiional imitation of the sounds in I nature. As for' what he calls eKclamations (mteljections), their

varia:tion, if we compare different languages, proves that we cannot

actually tn1k about motivation. And even if we admit that the two Icategories mentioned above :represent special situations, their limited

rromber and ml!rginal position in the language will not allow them in

be considered significant exceptions to the general rule. IAs pointed OIIt above; Saussure's postulation of the notion of arbitrariness sparlred a great deal ofcontroversy. The relation between the sounds that make up the word and the meaning that word has was [not, of courne, a Sllbject that the Swiss linguist analyzed for the :first time. As early as in Plato '8 dialogue Craty/()s. Socrates asks if the names we use for things are selected so that they correspond to the fnature of the things they refur to or if this selection is entirely

10 "A n'imp""" quelle opaque et si baut que nous re:tllDmiOnS, 1. langue apparaIt tonjattrS comme un bCtitage da l'epoque precedenre. L'acre par leque], a un II moment donne, les IIlDIS senUcm distributs Rill( cllos.., par 1_1 un e

J

J

J J

arbitrary. While Cralylos supports the first point of view, Hennog=es, SOcTates' othednterlocutar favours ilie second one. The philoSQpher do"sn't. explicitly opt far any offu" alternati:ves, offering :instead a number ofmore,ar less doubtful etymologies. The notion of arbitrariness ali lack. of motivation postulated by Saussure was subsequently refined by 1i.nguists. It was arguedthat there is some sort of motivation in the .use of any lexical item. What we should talk

;I about w :rather fJ:urt this motivation can have dtlferent degrees of obviousness. A dtlfenmce was established between absolute. or... cxttmlal motivation, lying in the very nature of the acoustic image tbat somehow suggests the meaning of the word, a case illustrated by~. onomatopoeic words and interjwtions (Saussure's objection that there ~

1 are differences between various J.an.,auage5 - in order to express pain, for instance, a Romanian will say "au" while an English.man willsay

1

rather "ouch", the word that is used to iuuMe a dog's barking is- "ham" in Romanian, "woof' in English and "oua" in French - was

dismissed ali irrelevant) and relative or inl:=al motivation where the:....

meaning of the word can be analyzed starting from its structure in which case we can.' talk about morpho\ogk.al, semantic or phonetic motivation.The morphological structure of derived and compound

;I

-

words ean thus offer a clue to their meaningY Since we know, fur instance the meaning of both the word home and the suffix -less we will be able' to analyze the meaning of homeless. Compounds will be a ~ more difficult ease smce some of them are, indeed, semantically t:ranspar=t _ everybody will be able to understailll that a taxi drive:r is someone who drives a taxi, and a hothouse is some sort of cOl1lltruetion where a high temperatore is preserved inside' - oilier

....

:I compounds are opaque, b1.nce obvioUJlly" hot dog is not a dog that.is hot and a Ted herring is not a red fish. Even in such Cases it CIlll be argued that an eiy!llological analysis of the word can. lead to a

...

successful inl:crpretation of the word -as it will uncover the semantic

:'. J "'" II It should be noticed, however, tba! such analyses account for Ihe W:rj ill

which tile word was fanned (derived or compounded), but Ihe ultimate componen:ts of tho word free n:i.Orpbemes or:affixes - still remain umnotivated in

,# tbe Sanssuriall sense.

:1 2(1

changes the word underwent. Thus a word tlmt was fozmed on the basis of a metaphor becomes a fossiliZed expression and the initial motivation is'lost. Phonetic changes also alter the structure of words and again initial motivation is lost It is difficult for an Englislunan who is ignorant of the etymology of the word to link the: modern English word rely to the Latin Tcligare and to see that it is related to rally too. The motivation is also lost jfsome words are not inherited or

. are lost in the history of a Ianguage. Any Romanian will understand the word oie:r (,shepherd) as referring to someone taking care of oi (s/wep, plural), while its synonympticurar will not be easily related to the Latin word pecora (sheep) as tins word was not inherited in Romanian. An inteJ;eSting l?ut somehow opposite caSe, will be thafof the so-called "folk etYmologies" which represent attempts of a linguistic' community to' aSsign some sort of motivation to an otherwise opaque fomJ by modi:fY.ing- the phonetic structure of the word and making it similar to other words in the language. Thus the French word ecrevisse becamtl cray fish in English. An interesting ease ill the ung:ramriIatical :furm tran(s)versa instead of traversa (cross), a word ofPrench origin (traverser) in Romanian. 'Ibc word ill mispronounced by analogy with other Latinate words in Romanian where the pre;fix trCUlS (across) appears. However, in this way the real origin of the word is uncovered, smce it was initially formed from tran.~ and 1I1!.rSllS.

The second essential feature that defines the sign and that is IliscllllSlld by SaUSSU;;~ is the linearity of the signifier. By this, the Swiss lingmsts U!ldeJ::,lands ffiiifliOi:1i in articulatoy terms and in auditory ones the signifier is ebaraCnmzed by duration. This duration is unidimensional and is conventionally represented as a line including the successive mOrot::rl!s in time. It takes time to utter a word and it takes time to perceive and 'QIlderstand it. The producing and the analysis of the signifier are procesies that unfold in time, that are made up of successive stages: On the contrary, the signified is something of which we have an instan:taneons perception. We 'tan compare this to am perception of visual signs which is simultaneous and multidimensional. This is more obvious, SaussllTe argues, ifwe

21

tilink of tile written aspect of languages. Jwy writing convention is, indeed, based on tile principle of linearity.

Ferdinand de Sl!l.lS.rore's COUTS de linguistique, generaJe (published a couple ofyears after his death, ill 1915) by several ofhis students, marked a turning point in the historY of modern lillgu:i~cs . find his themy of tile linguistic sign (tilough ninch refined by otiler' linguists) remained a cornerstone fur all subsequent tileories of reference. It is partica.darly relevant for understanding the importance and role of phonetics amOng ofuer linguistic disciplines. As we sb.all see la1er, since phonetics is concerned with tile study of soullI:lS.inl doIllJl:in is clearly iliat oftile signifiant.

Another linguist's contn1iution to tile underntandillg of tile linguistic sign is particularly relevant for am discussion: the Danish linguist LOllis I:!jcl:mslcv, the most outstmding representative of tile glossemstic school. Hjelmslev, too, descnlle.s the linguistic sign as a binary reelity. He distinguishes between tile leveIs of

JII

J essential in Hjelmslev's theory. 14 Any significant unit or any fOIlll at

'I both the level of expression and the level ofcontent is thus defined by the contrast in which it stands with all the other units in the ~yslen1. Saussure had already llsed the teon diffirence when describing this

-

fundamental property ofall languages. 15

J ...

1.4. Language in the process of communication

Language is obviously the main system available for us, notII only for knowing the world and understanding it, but also for accumulating, ;,toring and communicating information. Language can- thus be understood as the main system we have ror communicating

among 11S. All the other SYstems of conveying infoImation are actually b-dSed Oil this essential, fundamental one, Communication by means of

;....

lan"uuage can thus be understood as a complex process actoally

:1 consisting of several stages. Ally act of communication basically takes place between two participants: on the one.band we haV~ the source of...

the information" the person who has to communicaJ:e something, the sc:nder of the message that contains the infonnation" and on the other hand we need a second' party, the recipient, the addressee of

...t

the message, the beneficiary of ,the communication act, in 'other words the p'erson(s) 10 whom the information contained in the message is addressed.

....

I t

Since the sender has to convey a message, and the transmission is to take place on the basis of a system of signs (a code), the :first

....

.. Coci no.. montre que h>s d,ux fonctifu qui collln!Ctent Ia fonetionJ semiotique: l'exptession el Jo conIImU enJrent dans Ie.rname rapport avec elle. C'est --

seulrunent en verin de 1. fonction sCmiotique ([U'ils roOsteD! et qu'an peut los designer avec precision comme Ia fi>nne du conlmu et Ia fmme de l'cxp.....ioll. De roOme, c'est en verta de 1a funne du conllIDU et de 1. forme de l'cxpression seulement qu'existent 1. substance du conrenu et Ia substance de fl>ll:pression qui apparaissoot ql.land on prPjette la iinme sur Ie sens, COlllIIle un filet tendu projelIeI

... son orub", sur Ul1e sndilce ininterIDmpuo. (1968:81)

15 "Tout ce qui precede ",vienl it

, "

: ~

,I

the semarmc encoding. Then the respective words should be given a fann in accordimce with the rules'ofgrammar. It is obvious that ifwe say "fue door WI2\' open" instead of "Ine door is open" we do oot convey the SIIlrul idea: The correct choice of fue approprlJ!:te tense funn pertains to the domain (lfmorpbology. It;. on the other hand" we say "door open the is", the fOIm of the message is clearly' ungrn:mma:tical since ,it blatantly violafes the roles of syntlUt (word order in the given case). ,Both morphological encoding and syntactic encoding can be considered as stages ofwhai we can call grammatical encoding. Once it has llI1 appropriate irnmmaticaI furm, the message has to be give!). a phoJ).etic sbape, in other words the ideas we have to

con~"y must be put into sounils. nus last type of encoding is called phomlogicaZ encodtng. Some linguists describe the translating of concepts irrfn words and the assigning of a phonological shape to fuose words respectively as two different types of articulation and they speak about the' double arliculation oj the Iang>Ulge. Thos, according to Andre Martinet, the first articulation of language will include the segmenting of the conteJlt level by a given languagesystem and the association of acoustic images to the concepts tbn.~ obtained. The minimal units for 1his articula:tion are the Words, having both meaning and a phonetic stroctore. The next articulation will imply the segmcntatio;' of the aconstic image into corxtrostiv'e units,

the phonemes. Though devoid of meaning themselves, these units

have the eSseutiaI function of keeping different words apart Martinet

convincingly argues that Iaognages display remarkable economy at fue

level of the Second articulation since aconstic images are decomposed

into' a limited mnnber of significant units, !he sfstem mak1rig use of

the latter's eJd:J;aotdinary combinatorlal possibilities instead of

associal:ing each acoustic image to a different unit 17

A&r the 1:zan8mission of the tneS.'lage, the recipient has to decode it, petfumring the same operations, but in the opposite order, IlB fust has to decode the message phonologically, then to decode it gJ;amn:latically and then. semantically, reaching thllS the actoaI infurmational contllllt

11 Martinet. 1970:13-15. These notions will be llUbsequeutly explained in further detaiL

Summariring, we can say that the communication process 'takes place 'ai:cording to !he'following pattern:

~~~ffue I , 1. Semantic en'coding oftbe

me.... age 2.Grammatical encoding of the message _ 3. PhonolOgical enroding ofthe message

II ~ I '

IRecipient ofthe message L Phonological decoding ofthe mess.age 2.Grammatical decoding oftne message 3.Semantic decoding ofthe message

~ TranSD:J..ission ofth~message

- 1.5. Language and writing

:-,' -The importance of language and of it~ stody fur the undetsJaniling of what essentially characterizes our very nature as hI/man Beings cannot be overestimated. By means of the language immans not only co=icate in a fuller and more efficient way thao any other species," but they are in fact the only creatures 1hst can tmnsrnit information from a generation to fue following ones. We have sO':fur discussed lin",auiatic colIlInunicaUon only in its oral form (whichiS; ofcourse, relevant for fue Sllldy ofphonetics), but for many thoilsanrui ofyears now human beings have communicated in a written fOmi as- well. While speech was probably an essential faculty that cl:laracterized ',humans from -the first stages of their existence as a di:!ierent species from !be rest of the arrimaI kingdom, writing

" Scientists have elCtensiveJy studied'various means !lJrough wbich anhnals communicate with one another. Though they have sometimes been loosely called ''Jan&nages", tile di:ffi:nmt modalities used by eer1ain species ofm!llll!llals or by birds in onlc:r to c:ommunlcate can in no way be cOI1lp>red to tb. complex lin"onistic system ofcommunication humans USe.

27

~_I -)' :...

1L [I ;J

J

:J

J J J J I

I

I

...

cl certainly appeared much laler in the history of mBrildud The earliest records ofhlllJllilL'l tr;ying to express their thougb!s inwriting date back

...

to only several thousand yea:rs ago. Even today there are many- languages that do not haVe a written folJD.. The "invention" of writing

was essential in. the process of '!rIlD1llllittig inforroation over gl;eatt distances, both ill spare and in t:i.ole. It played a,tremendous role in the development and evolution of hmnan: ci'Jilization as it is mainlY

-I through written recoxds that info=ation abo1;lt civilizations that have long been extinct managed to reach us. Sueh is the importance of .J

writing in modem t:i.oles that we tend to neglect its :relatively younger

J age. We forget 1;)lat many languages in the past and even at present were (or are) exclusively spoken lllld that writing is, after all, a secondary and relatively less important system of symbolization19in the absence of which linguistic systems can function very well. The prestige of writing is so great that the written furm of the word1

....

1 influences oW: Jllental representation of the word and we often tend to reverse the natural precedent speaking has over writing and to consider writing as being primOrdial and speaking only secondarY, ~ This is due, as Saussure points ont to the fuct that graphic symbols tend to make a more lasting impression on our -intellect than the ~ sounds we hear. TheY give ,UH the illusion of soliditY and permanence when, in reality, grapbic Conventions are by fur more superficial and

rI irrelevant fOf the basic :fual:utes of the Ianguage.2Q II ';'M;+- 1e

. 19 Our own. language, Rom:miarr is (somehOW ~y) an ellI1 in Romanlan daleS from 1521 but the historY ofthe hmguage itSelf stretclles baCkman:y =1""bero", that

iI r..:.: ,. The ex;unple of Rl:>nI3trim _ be again qut.ed. The Slo.voni alphabet

was used unti1lhe second' half of tb;> 19'" C Ro)Illlll alphabet was adopted. Turlrish abandOned its tra

their initial value and have acquired a phonetic character. Chinese pictogrnms and Egyptian hieroglyphs oirer such examples?l

ff we consider the cbronoln~ca1 evolution of writing we C811 notice a !J:1lIl3ition ;from rurect, more or less concrete systems of represetrtatiorui to increasingly abstract (thougll p~ca11y simpler) ones. The manner in which human beings tried tn convey their ideas . using graphic represeittations was futally vety rudimentary in. 1he beginning. The earliest syStems of writing (if we can ind~ speak of writing) were actually visual representations of what men saw. The scenes of11lmtil)g, for instance, painted on cave walls, ate considered 1he first a:tternpts of human beings. to give their thoughts a grnphic form. Later s1ag

I

,I

CHAPTER 2

J

J ARTrCUI..ATORY, jilCOUSTrC AND

AUDITORY PHONETICS. PHONOlOGY

J 2.1. Phonetics and phonology

J J Two tt:l:ll:ls are (often loosely) used to refer to linguistic

disciplines studying fbat part of the linguistic sign which de Sanssure called the acoustic image: phonetics and phonology. The importance

J of sounds as vclricIes of meaning is something people have been aware of for tliousiUIds of years. However, syst!lmatic studies on the speech sounds cmlji- appeared with the development of modem

-I sciences. The texin phonetics used in CO!lllection with such studies = from Greek and its origins CIlIl be Iraced back to the verb phOnein, to speak. in its tutn related to phone, sournt The end of the

-

18th century witnessed a revival of the interest in the studying of the somids of various language.q and the introduction of the term

1 phoflOlogy. The IaUer comes to be, bowever, distinguished from the -

former only more than a ce:ntUIy later with the development of structuralism which emphasizes the essential contrastive role of

1 classes of sounds which lire labeled phonBmes. Ibe terms continue to -

be used. however, indiscriminately until the prestige ofphonology as a distind discipline is finally established in the first balf of the 20th century. Though there is no universally accepted point of view about a

'I clear-cut border line betw"'-:n 1fu, respective.domains of phonetics and :1 ,,"

phonology as, indeed, we c>innot talk about a pbonological system ignoring the phonetic aspects it involves ana, on the other band, any phonetic approach shonld take into accouut the phonological system

....

r, thai is represented by any language, most linguists will agree aho LIt SOIIW fundameutal distinctions between the two. '-I 33

Phonetics will be almost unanimously acknowlec!ged to be the linguistic science which studies speech sounds: the way in which they are produced (uttered, articulated),' the way in. which they are perceived, their physical chiracteristics, etc. Therefore, it is these more "palpable", measurable aspects of the phonic aspects of. language thet.constitote the domain ofphonetins. On' the other band, it is obvious, however, even for those whose perception of lil::i",anistic phenomena is rather of an empirical and not of a very scholarly kind, that when cornmunica:ting verbally, though they are producing a wide vanely of sounds, people are actna1ly "aware" .of using a comparatively drastically limited set of sounds, in other words that . they tend to disregard the obvious (mOre or less important) differences between the way in wbich sounds are uttered and have in mind only classes of sounds that perfurm a certain function in language; From this new perspective, it is not the sounds as such that are impozt,mt. but rather the role they have in linguistic communication. As we shall see later, diffi:rent languages operate different distinctiOIlS and stroctorc in di.fferont ways the more or less common stock of sounds thet can be round in various idioms. It is precisely Ibi!> ru,-pect of sounds that is of interest for pho7lf}logy, which. is !bus understood to study not so much the sounds as such, but rather clas5e$ of sounds that have a certtrin function in the stroct-ure of a given Iimguage,l This distinction will be further analyzed in the chapter dealing with the phoneme..

We have alreadY said that phonetics is concerned with various aspects relev-dllt fur the physical characteristics of sounds. Several lmmches of phonetics can further be distinguiShed, depending'on the narrower domain of interest of the respective field. Thus, one of the most important branches of phonetiCli is articulata", phonetics which. studies the way in which h11lIJJll1 beings articulate or utter the sounds they make use ofin verbal communication.

, Note llJat what we refured 1

J J

physiology ofthe respective vital processes. The .!unll!! i!l!e pair organs, situated inside the thomcic cavity (the chest). They are farmed of three, respectively two spongy lobes (the left lung is smaller because

I of the vicinity of the heart within the thoracic cavity). The capacity of the lungs (that is the totat amoWlf of air that they can conlain is:,of about 4500-5000 cm3 (4.5-5 litres) in an adult person, the capacity

-'

beirig glillerally slightly superior in the case of male persons. The so-

J , .

J called vital capacity (that is the mrucima1 amount of air that can be exchanged with. the environment during breathing is of about 35004000 em'. In other words, we can never completely empty our lungs of air during expiration. During nonnal breathing, however, only about 10-15% orthe vital capacity is used, that is the qua:nJity of air

"I that is exchanged amoWlfs to about 400-500 em'. The act ofspeaking requires a greater respirato;:y effort and consequently the amount of air increases to up to 30-80% of the vital capacity (30-40% during

- expiration and 45-80% during inspiration). VariatiollS are due to different position of the body, to the quality, quantity and intensity (lowlm:ss).f-the sounds we art1

of the angle, oriented posteriorly, varies with the sex.. It is a right angle in men (90) while in women it is. 120. The angle is more visible, because more acute, in the farmer sitnation and the cartilage is popularly known as "Adam's applc". Posteriorly, each of the plates bas twa hams (an inferior and a superior one) called: cornua, thTOl,lgll which the thyroid cartilage is COlll1ected with the ericoid one. The joint. that the two cartilages fonn, resembling a sort of hioges, allows the cricoid one to moveanteriorly and posteriorly with respect to the

. .

thyroid one, 1frug controlling the degree of tension in the vocal cords. One of the main functions of the thyroid cartilage is to protect the larynx and particularly the vocal cords. The cricoid cartilage is made of a ring-shaped structure, situated Ill1teriorly and of a blade situated posteriorly and represents the base of the larynx, controlling co=unication with the trachea. On top of its blade, on the left and right side respectively, another pair of cartilages are situated: the arytenoid ones. The last important cartilage in" the process of phcmation or speech production is the epiglottis which is a spoonshaped"cartilage also playing an important role in keeping the fOod away :from the respiratory tract. It is between the arytenoid cartilages and the thyroid cartilage .that the two vocal cord. mentioned above stretch. The mcal cords are each made of a so-called vocal ligament and a vocal muscle. They are covered in mucous membranes or "kin folds also klIDwn as the vocal folds. They connect the lowl'J: part oHhe thyroid caitiJage to the anterior part of the arytenoid cartilages" The" opening between the folds and the arytenoid eartilages" represents the glottal aperture, more co=only called the glottis. The length ofthe vocal folds varies with the age and the sex. They become lon~r at the age of puberty and are longer in men (24-26 =) than in w"Omen (17-20 mm). During"breathing , the two folds part, letting the air come into the larynx or go aut. During phan;mon they come closer, having an important role in establishing some of the main chm:acteristics of the sounds we articulate. By the pretty complex action of adjacent anatomical strUctures (the cartilages described above and a number of laryngeal muscles) the two vocal cord" ean be brought together or parted. They thus interfere to various extents with the outgoing

:H~

;"~

I

airstream. They Clll1 obstruct the passage completely, as in the case of the so-called glottal stop (see below, when a detailed description of consonants is given), or their participation in the uttering of a given I sound can be mjnjmal (as in the case of trumy hissing sound~). The rapid and intenni:tten:t opening and closing of the vocal cords, which results in the vibration ofthe tWo orgdllS, plays a key role in aile ofthe ] most important phonetic processes, that of voicing. Thus, vowels and vowel-like sounds, as well as a number of consonants, are produced with the vib,,!tiori of fue cords and are consequently v('Jiced. The ] absence of vibration in the vocal cords is characteristic for voiceless obstruents. (More details about the process me given the following Chapter). The amplitude ofthe vibration is also essential for the degree I ofloudness of the voice: thns the intensity of the sound that is uttered

depends On the pressore of the air that is !""P"lled. The rate at which

the vocal cords .vibrate has also importabt consequences as far as the I pitch of the voice is concerned; this is closely linked to the preSsure

exerted On the vocal cards. When" we 'produce acute (high-pitched or

shrill)) sounds the vocal cords come closer to each other, while duting I the articulation of grave sounds" the vocal cords leave a greater space

between them. (Further details will be given below, when tbe acoustic

characteristics ofsounds are discussed.) I

The Jlext stop on our way along the vocal tract is the pharynx,

an organ siruated at a kind of cros.

I ] J

ro.'terior ofthe mO]l~ ~ontinues with 1;he velum.into the rear part ofthe

mouth. The'latter's position at the back of the mouth can anow the

airstream to ,\0 o";t through either the mouili or the nose or through

I) both at the slii:QZ :time, Thus, if the velini is raised, blo!ing the nasal cavity, the air is directed out through tfu: mouth and the sounds thus .

J

J

, produced will be oral sounds. If the vclum is lowered; we can'

articulate either nasal sounds, if the air is C}..'Pe1led excluSively via the

~a.'al cavity, or nasalized sounds if, in spite of the lowered poSition of

the velUm, the air is still allowed to go out through the mouth as well

as through the ~ose. If we nip our nostrils or if the nasal cevity is

bloc:kl'\d ):lecause of a cold, hay rever, etc, we can easily notice .the

[J

importance of the nasal cavity as a resonator and the way in which its

blocl\ini affects nonnal speech production. The distinction nru;a1/oral

is essential in all languages and it will further be discussed when a

detaile

the upper teeth and their position ( rounded or &1JfWl) is also relevant for differentiating b

I J

consequently talk about its peripheral and its central part, respectively. We shall have a closer look at both these processes and try 1:0 Show) why they are both clearly distinct and at the same time they are ciOHely,related.

.. Before the sounds we perceive are processed and interpreted J by the' bra:irl; 'the first anatomical org-.m thejr encounter is the ear. The. ear haS 'a 'complex structure and its basic auditorl funcl:ions include the perception ofauditory stimuli, their analysis and their transmission1 further on to ilie brain. We can identifY three components: the 'jut!!", the middle and the inner year. The ourer ear is mainly repIeStIDtOO by the auricle or the pinna and ilie auditory meatus or the ourer ear canal. ] The auricle ,is ,the ouly visible part of the ear, constituting its outermost part, the segment of ilie organ projectingQutside the skulL It does not play an essential role in audition, whick-is proved, by the] fact that ilie removing of ihe pinna does not substantially damage our auditory capacity. The-auricle rather plays a protective roklo! the =1: of the ear aud it also helps us locilize sounds:'!fhe meatus, or the onter ] ear canal is a tubular structure playing 11 double roll::: it, too, protects ilie neltl segments of the ear - particularly the middle ear - aud it also functions as a resonator for the sound waves that enter our auditory ] system:. The middle ear is a eavity wi1hin the skall including a number of little anatomical structures that have an irDportmt role in audition. One of them is the eardrum. This is a diaphragm or membrane to [J which soOOd waves are directed from 'outside and which vibrates, acting as both' a filter and a trnnsrnitter of the ineoming sounds. The

J middle ear also eorrtalns a few tiny bones: the mallet, the anvil and the stirrup. The pressure of the air entering our auditory sjstem is converted by ilie .vibration of the membrane (ilie eardrum) and ilie

J elaborate movement of the little bones iliat act as some sort of lever &ysterii into mcchanical movement which is further conveyed to the oval window, a structure placed at the interfuce of ilie middle and

, inner ear. As pointed ont above, th middle, ear plays an, important [I, [J We should not forget that the em (more eXllCtly llle inner ear) also plays an essential part in our body's capacity of keeping ils balance. As this functioIl of the ear obviously lies out'1ide tbe scope of!his book we are not going to discUSS it.

44,~

protection role. The mnscles associated with the three little bo,ne5 ~ntioned above' eontmet in 'a reflex movement when sounds having a too higll intensity reach the ear. Thus the impact of the too Loud sounds is reduced and the mcchaniSlll diminishes the force with which the movement is transmitted to the struchrres of the inner ear. It i, in the middle ear .too, thiit a natrow duct or tube opens. Known as the Eustachian tobe it connects ilie middle ~ to, the pharynx. Its main role is to act as an outlet pcrruitting the air to circulate between ille pharynX aud the ear, thus helping preserve .the required a:mount of arr pressW:c inside the middle ear. The neltl segment is ilie inner ear, the main element ofwhich is the eochlea, a cjl.vity filled with liquid. The inner ear also includes the vestibule of the ear and the semicircular' canals., The vestibule represents ilie central part of the labyriuth of the ear and it gives aCcess to the cochlea The,cocblea is a coil-like organ,

~, ". .

100king1iJre the shell ofa snail At each oEthe two {--nds of the cochlea

there is an oval window, while the ~rga.ti itself eontains a liquid. fuside

the cochlea there are two melllbranes: the vestibular membrane and

the basilar mCll1brane. It is the latter that plays a central role in the act

of audition. Also essential in the process of hearing is the so-called

organ of Corti. inside the cochlea, a slruetore that is the real auditory

receptor. Simplifying lllot, we Can describe the physiology of audition

inside the inner ear as follows: the mechanical movement of the little

bony structores of the middle ear (ilie mallet, the anvil and the stirrup)

is transmitted through the oval window to the liquid inside the

snaiHike structure of the cocblea; this causes the basilar membmne to

vibmte: the membrane is stiffur at CJ11e and than at the other, which

makes it vibrate differently, depending on the'pitch of the sounds that

are received. Thus, low-frequency (grave) sounds will make vibrdte

the membrdllc at the 1es8 stiff (upper) end, while high-frequen.cy (aruIc) sounds 1vill cause thc lower and stiffer end of the membrane to vibrate. The cells of the organ ofCorti. a.highly sensitive structure hecause it :i:n.clndes .many 'qiliate cells that detect the slightest vibrating movement, convert these vibrations into n.eural signals that are tranmnitted via the auditory nerves to the central receptor and controller ofthe en1rre prOcess, the brain. '

45

The way in which the hUlIlaIl brain processes auditory infi:mna:tion arul, in general, the mental processes linked to speech P"Xception and production are still largely unlmown. 'What is clear, however, regarding the perception of sounds by man's auditory system., is that the buman ear can only hear sounds having certain amplitudes and :frequencies. If the amplitudes. and frequencies of the respective sound waves are lower thM the range p=-ptible by the ear, they are simply not beard. If, on the contrary, they are higher, the sensation they give is one of pain, the pressure exerted on the eardrlllTlS being tuo great. These aspects are going to be' discussed below when the physical properties of sounds are aualyzeJ. As to the psycho logicai processes involved by the interpretation of the sounds we hear, our knowledge is even more limited. It is obvious that hearing proper goes haud in haud with the nndersf:a:t1ili of the sounds we perceive in the sense of organizing them according to pattemll already existing in our mind and distu"bntiog them into the famous acou&llc irrlIIges that Saussore spoke of. It is at this level that audition proper intermingles with psychological processes heMuse our brain decodes, interprets, cla.sifies and arranges the respective sounds according to the linguistic (phonological) patterns already eristing in our mind.7 It is intuitively obvious that if we listen to someone speaking an uriknown language it:will be very diflicrdt fur US not only to understand what they say (this is out of the question given the premise we sfmted from) but we will have great, often insurmountable difficrdties in idcntH'ying the actna1 sounds the person produced. The immediate, reflex reaCtion qf our brain will be tu' assimihrt.e the respective SOUllds to the ones whose mental images already exists in our brain, according to a very conunon cognitive reaction of hUlIlallS

Listening ~n is, ofl'e.- oJ!, an essential part of our ttlalllCring C and we will be able to nnd".."tand 111. word the speaker meant to utter in spite of its actual faulty pronnncigtion (see, millO, note 5 above).

]". that always .have the tendency to relate, compare and contrast new

information to already J,::nOWR infOl'Ill!ltion. Our discussion of the

phonemcin a subsequentchapter will analy"~ this in further detaiL ]

;:

2.4. Acoustic phonetics ] The branch ofphonetics that studies the physical parametres of

speech SQunds is called aCQustic phonetics. It is the most "technical" of all disciplines that are concerned with the mtdy of verbal ] commLttJica:tion. The data it bandles are the most concrete, palpable, easily measurable ones that can be encountered in the domain of phonetics in general.' The most impo:rt.ant principle of physics on il which verbal oornmllnica:ti911 is based is that VJ.btating bodies send

waves that are propagated inthe env:iroi:ty:teut. Our articlliatory organ.~

produce a number ofvibnrtiOl1$; these VIbrations need a mediUlIl to be I tr.a:r:!Jil.nitted through. The medinro through V\lhich speech sounds travel

is tJ.

J

J movement would continue for ever. However, as the vibrating body is 1 surrounded by its movement is'1:!:llIlSD1itted to the air molecules around. that vibrate accordingly. The vibration of1he pendulum or of - the prong of the tu:ni:og furk can be represented graphically by a

sinusoidal curve. The vertical axis or the ordin;rt:e will measure the amplitude or intensity of the SQUDd. while the. horizontal one, or the1

..

1 abscissa will measure the duration in time oftile vibration. . .

If the distance from tile point of rest is greater, we say the amplitude of the vibration is higher. This is related to the aruount of

... energy that is traIllilIlitted through the air by means of the resPective sound wave. The higher the amplitude i~, the louder the sound. The1 conventional way in which we refer to the intensity or loudness or

'I .- ampJi:tude of sounds is tIlat of using the decibel scale. The decibel

scale does not express the absolute intensity of a sound, but the ratio between the intensity of a sound and a reference intensity. Thus, ifwe

...

1 want to compare the intensity of two sounds, we take the logarithm to the base 10 of their ratio aod multiply it by 10. For instance, ifa sOlIDd is 1000 times more intense than alJother, it means 1:baf 10 bas to be

J *, raised to the power 3 1JJ get the ratio between them. Ifwe multiply 3

by lOwe get 30, therefore the difference between the two sounds is of 3a decibels (dB). If a sound is a billion times more intense than another, this means that their ratio. is 10 raised to the power 9, so 1he dilference between them is of 9 multiplied by 10, that is 090 decibels (dB). Tho reference value fur the decibel scale is the standard intensity1

.... of a sound which has a fixed value close tothe audible limit of sound. (Ihis'value is .10,16 watts per square centimetre). Therefore, ifwe say1 that a sound is 40 decibels it means it is ten ihousaod times more

- intenSe than ihe standerd reft:rence value. A complete' movement, thin is one starting from tile initial;. point going as far as the IJlllXimum amplitude, ,thw back to the point

of rest aod beyond it to the maximum amplitude in the opposite -

direction and fulally back again to 1he point of reat is called a cycle.1 lhe higher thc number of cycles per unit of time (second) is, the 1 0..; higher thefrequency of the vibration is. The time it takes for a cycle to

be compkted is called the period of the vibration. Frequency is

"';

t 48

measured in cycles per second (cps) or Hertz. Sounds having a com;timt period (in other words sounds displaying a regular vibration) are called periodic sounds. The typical example for this kind of sounds are musical sounds. However, in the case of other sounds, snccessive periods vary and these sounds are called aperiodic. In reality, periodic vibrations are seldom simple, the vibration being of a more complex kind than that rL-presented by the simple sinusoidal wave (or sine "I'II1;lve) deseribed above. A vibrating body oscillates or vibrates at various intensitie.', the ensuing vibration of the entire body being a wave that is not si]rusoidal and will differ from any of the simple sine waves of which it is the result. The sinnsoidal components of aoy complex periodic sound are called the harmnnics of the respective sound. The higher harmonics are integral multiples of the 10WCb1: harmonic which is called 1he fo,ndamental frequency or the fondamentaI of the resPective sound. Thus, if a sound has as its fundamental frequency 200 cps and one of its higher harmonics is of, say, 400 cps, we say that.;the latter is the 2nd harmonic of the sound since it is twice bigher than the fundemental. A harmonic having the frequency of 800 cps will be 1he 4th harmonic of1he sound, as it is four times higher than the fundamental. We should always specify therefore, in the case ofperiodic sounds, which are the frcqncocy and amplitude of its funda1:nental and of its higher hru:monies. It is also important to note that though the various mtes of vibra1ion will result in a given timbre (tonality) of the sound. which is different from any of 1he harmonics, it will always be the fundemental that es"entially defines (give., 1he quality of) a given sound. This kind of specification that include., the fundamental and -the hru:monics of a sound is called 1hc spectrum oftlie respective sound.

An essential feature of any sound is its pitch. Pitch is, roughly .-peaking, the way in which we perceive the frequency of a SOUlld, it in other words the pc.:rceptoal correlate of the frequency of that sound, We can say that the higher 1he fundemental frequency of a sound will be, the higher the pitch of fue respective sound is, or nuher that we perceive the sound as having a higher pitch. This correlation is not, however, linear as there is not always a direct proportionality between

the frequency of a sound and our perception of that frequency. Pitch has a very important role in intonation as we shall see later. Pitch differs a lot from one speaker to another. Women, for instance, have shriller voices than men, therefore the pitch of their utterances will he higher.9 How is it then that we recognize a sound as being "the same" even if it is pronounced by persons whose voices have very .different . pitches? The answer is that though the fundamental and the number of harmonics differ, obviously, in the two cases (the one with a lower pitch having a lower number of harmonics) the shape of the spectrum of the two sounds is pretty much the same in the sense that the harmonics with the greatest ainplitude are at about the same frequency in both cascs. While vowels and sonorants have spectra which resemble those of periodic sounds (of the kind mUsical sounds are), obstruents, and particularly the voiceless ones, are aperiodic sounds. which malces them pretty similar to pure noises.

Three are them the essential acoustic parametres that characterize a given sound (a sound having a certain quality): its ampliIude or intensity - that we perceive as loudness; its frequency, that we perceive as pitch. and its duration. A given sound. therefore, say the vowel lei, can be pronounced with vru;ious degrees of intensity, the amplitude varies therefore, but fundamentally the sound is the same. In spite of frequency variations (that we perceive as variations in pitch) in the pronunciation ofthe above-mentioned vowel by different persons, we will still identify the "same" sound. We can also vmy the length of the vowel and we will still say that the sound' hasn't fundamentally changed its quality. The anatomy and physiology of both the articn1a:tion and audition processes draStically limit the range of sounds that we can produce and perceive, respectively. In oth~r words we cm only utter sounds within a eertain range of intensity and loudness and their dmation is also limited. Conversely, our auditory system is able to perceive and analyze

9 The frequency of vocal cord vibration ranges, generally. between 80 and 200 Hz in men, whlle the vibration of women's vocal cords can reach 400 Hz (see Ladefoged. 1975: 163)'

sounds whose frequency and intensity are situated between certain values and whose duration is limited-

The vibrations of a body can be transmitted, often with a higher:amplitude, by a phenomenon called resonance. Certain bodies have the property of transmitting vibrations in this way and they are called resonators: It is enough to think ofmusical instnnnents and this physical proceSs becomes clear for everybody. Ifwe take a violin, for instance, the strings play the role of vibrating bodies, while the body of the instrument acts as a resonator. And this is trne not only for string instruments, but for wind instruments as well. Ifwe take a flute or a bassoon, we shall easily see 1:hi.t the air that is pushed into the instrument when we blow:it makes vibrate the air already existing inside the instrument and the body of the instrument plays again the role ofresonator. .

A similar process can be witn~ssed in the case of speech. Remembering am description of the main articulators above we shall again mention the glottis as the first essential segment of the speech tract that shapes the sounds .tha\ we produce. The vocal cords have the role of vibrating bodies while tbe pharynx, the oral and the nasal cavities, respectively; act as resonators. The versatility of these cavities (notably the oral cavity) that can easily modify their shape and degree of aperture, the mobility of the tongue and the complexity of the human speech producing mechanism enable human beings to articulate a remarkable variety of sounds in terms of their acoustic :features. The initially weak vibrations of the vocal cords, having a wide range of frequencies, are taken over and amplified by the above mentioned resonators. The amplitude and frequency of the sounds that ate further transmitted by the resonators depend very much on the size and shape of these resonators. Resonance does not characterize, however; only cavities that modify the acoustic features of a sound. Vibrating.bodies themseives.are characterized by various degrees of resonance:cResonatws can. amplify or damp the fonnarrts of the given sound. by :enhancing or suppressing various frequencies. This accounts for the wide -.variety in the parameteres of sounds different human beings are able to produce. Each of the features of the

50 51

J

J

J

J articulators ofan individual has an impact on the types of sounds that individual utters_ The musicality ofthe sounds that we producc-largely depends on the characteristics of our phonatory- system, too. Vowels, for irist"'lce, have distinct and constant patterns of resonance (the resonating- cavities assume certain shapes whenever a given sound is uttered) and thus we Can always recognize the .respective sound by its

-'.

distinctive mark. The various positions of the soft palate will di:rect the air through either the onu or the IlIl.Sl!l cavity or through both offuem.

-.

This will give the sounds We produce a nasal or an oral character. As pointed out above, the shape and degree ofOjJ describe the femures. More detailS will be givOJI mthis book ill the chapter discussing di'/tin

I ,~

~I_.

(though arguably so, since we can hardly speak about a unique 2;5. SynchrOniC, diachronic, comparative phonology

. We have SO fur exatriined the interest of phoneticianS in what may. be called the production, the perception and the physical characteristics of sounds and we have briefly presentee! various doro.;iio.s' of phonetics dealing with the respective dam: articulatary, aud:il:ory and acoustic Pboni:rtics. LangtlllgeB, however, are not given once for ever and they are sUbject to change as all h= things are. Of course that changes affecting a certain language are not easily noticeable over a shari; period. of time and if we want to collect relevant data regarding these phenomena we often have to :refer to periods of one or severa} centuries. Pronunciation changes too and though we do not have recordings of the way in which people spoke centuries ago, specia1L

I

I

I

[I

J

J

]

II

]

J

]

]

]

rJ

FI

level and the kind of English spoken by a certain person often illustrales his or her educational and social bacl

repre~cnts an important index to lhe social and educational background of the speaker. This largely accollIJis :fur lhe survival of RP 'as the standard pIOllunci1ltion Ilf the ilmguage in spite of its stati.Stic insignificance descnoed in the not" before. Wilhin RP itaeif, however, three main types can be di,;

I

)

)

J

)

J

]

J

]

[)

II

[J

fJ

[)

rl

,the words of the language as any correspondence we might be tempted to establish between :the two can prove utterly misieading,14 Suggestions have .1)een made to simpli:fjr English orthography and "tune" it to :the prommciation of :the words. It is precisely tile m.1raordioo:ry variel)' of the language mentioned above :that seems t!l be, however, one of the major obstacle.~ in this direction, as; it has been argued, spelling T

year) or the :noun type is a hOJIlOnym of the verb type as they are hoth homophones and homograpb.q, while pairs of words lJke pray aod prey, meat and meet, sow ~d sew, will only ,be homophones but not genuille homonyms as'they are not also bomogrnpbs. We can come

,acro,~ 'the opposire situation, Wl:w.n two words are homograpb.~ but are pronounced differently: e.g. row (or;;b;irn)'aDd row Cqum:rel)bo~ (the

"weapon) and bow (the 'synonym of bend); $OW the verb and sow the female pig.

'CHAPTER 3

THE SOUNDS OF ENGLISH. CONSONANTS AND VOWELS. AN ARTICULATORY ,

CLASSIFICATION AND DESCRIPTION OF CONSONANTS. , DI.STRI~UTION.ACOUSTIC CORRELATES

3.1. Consonants and Vowels. Traditional distinctions. Chomsky and Halle's SPE definition

The previous chapter has provided a brief description of the

human phonatory system insisting on the main articulatory organs and

t;lifferentiating between passive and active articulators. Articulatory

phonetics has been defined j$JhII.Lbl@~9LOOmletics thatJl.l.Jldies.1hf'

soundS of a language from the point of view of theiJ:)U1LeolatioI.l....gf

the manner in which they come to be produceg, utttrred !:1Y the l!P.eliker.

The neXt ChliPters oftlils book will giv.;:u;;;description and ait~pt-a '

.,

classification of the sounda ofEnglish in articulatory terms, presenting also some acoustic correlates ofllie major classes of sounds.

When tJyjng to describe the, sounds of English - or of any language for that matter - one should, start with the ttaditional distinction between two major classes of sounds: v(JWels and consonants., respectively. There is, of course, no universally acce;pted definition" for ~ither cla~s - is there any subject upon which grammar:ians Will' agree, after all? but we can resort, at last for the beginning, to etymology, to explain what people - in an intuitive rather than scholarly manner -,have always under.srood by the two concepts. The woro vowel comes from. the Latin woro vocaiis, which ic its turn derives, from vox, vocis, meaning word, voice, (cf. also Rom.

6" .1

----

J

J

t< ---.~ vocaJif) In other words, we always percelve Ivowels', as sound

\~.~". -intimately relaied to the feature ofvoicedness: a vowellS a sound thai

J J must be produced with vocal cord V1;'ratio~: '&t tlrls'is a featuii-tri.iii

characterizes some non~v~~io sOWldS- is something thai will be discussed a little bit later. On the other band, the word ~~ su~llts - again, on a strictly etymological b/liiis - thai the'~ve

iJ ;~und doesn't have an articulatory autollomy, 0:;:, to put it differently, that it pas to sound tQ&-etber (La:, COnB017L1llS, present parrticiple of consonare cf also Lat. consOM, Rom. consoanli) or. ~ounccd in. association with other SOUllds. This iH again somerhin,i(tlia-Cwe are somehow intuitively aware of. or at least we were taught thai this was~I the case as early as during our first language classes in primary

....

schooL Thai this is a definition that causes some serious problems is again sometlring thai will be soon discussed:1

As pointed out before, these etymological references are notJ :1

very helpful in llllderstanding the true nature of the dlfferences betweOll fhe two classes of sounds. Though always voiced" vowels are by no means the only voiced sounds in a l!l.l1llU"ge. On fhe other hand,

,-sounds that don't have consonantal features, may very well be

;1 pronounced together with other sound.~. Further difficulties are created by fhe ambiguous nature of certain sounds that have both vocalic and ....

1 conson.autal features.

TIle soroewhaJ: intuitive criteria (seldom explicitly expressed, howeveri had to be replaced by systematio and consiStent attempts at

....

1 defi:n:ing the true nature ofthe di:ffurence~ between the two classes.

According to Ferdinand de Sau:;sure, what distinguishes vowels fram consonants is the higher degree of aperture of the oral

....

1 cavity. From an articulatory point of view, the two classes are not, however, essentially di:ffurent. It is only from an acou..otic point of view that distinctions are relevant; the laryngeal sound being

- amplified by the oral eavity that Dmctidns primarily as a resonator in

rl... I KOllllcth Pilre notes that "frequently fur desGripiWn ofsin~ langtUlgt'" 1he

;1 division is assumed, with no attempt In defiM it. Tl)e distinctiOIL is ofhm p~ as ifit were cl=rcu~ with every solllld belonging to one or the otlrer oftl!e groups" ",~

(1943: 66)

64'I

I

I

./ !I, j I

i

the .case of vowels, while in the case of consonants it r,,-duces the resonance of the laryngeal souru:ls, It noise-like effect being produeed by fhe In:tercession of oral articulatOrs.2

LeOl1llTd Bloomfield defines vowels as "modifications of voice-sOlmd that involve no closure, :fi:ictiDn, or contact ofthe tongue or lips" while conso1ll'll1tS (thai include stops, trills, ~pimnJ:s, nasals and laterals) are "the other" sounds. Bloomfield i:leplares the way in which the twa labels, vowels and consonants respectively, are used and argues that in. the description of individuBl1anguages it is convenient to use the terIns in a cliffurent w:ay and to supple:n:tlmt !his distinction made in artk.-ulatory 1etms. He suggesi.s that the distinction should be refined by adding two more classes: sonan/s and ;-emivoweL,', (1935: l02) ,

Arguing in favOlU' of a slrict delimitation betweL"Il fhe phonetic (articulatoJy and acoustic) descriptions of souru:ls and their phonemic, contrastive value in a given context,; J'ike' distinguishes between contoid and vocoid sound.q, a division exclusively based on phonetic charaeter:i:>1ies thai parallels the distinction consonants/vowels iliat are "categories of sounds, not as deten:nined by fheir own phonetic nature, hut according to their. grouping in speci:lic syllable contextual functions", . Aecording to tlris interprataiion we can talk about univernal, purely phonetic fua'tores of conroids aurl vocoi ds respectively, while each particular langllflgc (phonological system) will dclineate itS own classes ofconsommts and vowels.

1. La fomml. d'une voyell. est ."a,c!:fmlellt comparable a celie de n'irnporte

quelle COIlSOllil. s= Au point de we de "articulation buccal", il n'y a pas de

dJstinctian a:fitire. Scul I'cffot acOIllltiqUll est difierent. Passe 1m certain degr.

d'aperture, la bouche .tbru:tiom:te principalc;twmt comme resona:teUr. Le timbre dtt

son l..-ynge "Ppi1fldt pleineinen1 et la'bruit buccal.'ei!iu:e. Plu!! Ia bouche 50 ferroe,

plus Ie son bIIYllg&est:intercepti. (1965: 75)

'"A phonetic sYstem should be able, wiJirin the limits ofthe aecnracy and

finesse of its articulaWty, acoustic, ru- imitation-labo! procedures, to d"scribe my

sound in isolation., or in: nanse;n.sc syllablcs, or as cut from: the continuum of speech, without the necessity ofreftu:ring to othct sounds in 11m context to :find crltz.ria for its classtlic:ation. A phonetic sci""ce should be .ble In define IIIld de$noe its own uriits by its Oml dattt. If1he phanetic.ian ;fj,,;t dolilnits supposed mticnlatory classes by phonnmlc team.res, haw can be thon dc=ibe the phonemes with articulatory mel:bods? Any such a!tcmPt presents a Vicious circle of phonemics 10 phonetics ID phonemics, with the phonetician stllrting at phonmmcs.n (1943: 78)

65

Many conttmlponuy liDguistic. studies follow C'homsky and Halle (The Sound Pattern of English, 1968) in po~'tulatiDg the fact that the mmn distinction hetween vowels and COIlSOrumts consists in the .filet that while we utter a vowel the outgoing 8irl:tream does not meet any major ohstacle or constriction in its way from the lungs out of the mouth, and the articulation of the sound allows spontanOOtlS voicing, whereas the articulation of a consonant always involves some kind ofblocking of the airstream.4

Once we have decided that consonants are sounds that involve a smeture (llllmlwing, wlrich can sometimes lead to a complete Obstruction) of !he vocal tract, we will easily notic.e that what we have just decidedto call consonants are far from being a homogeneous elass. On the other hil!ld, it is obvious that consoruu.rts will be more readily described io articolatory terms than. vowels since it will be definitely easier to poiut to the precise organs involved in the process of articulation and to the place where the above mentioned constriction takes place.

3.2. Criteria for consolllant classification. Vocal cord vibration. Sonority

Traditionally, the three basic criteria llsed in the articulatory descriptioo of a sound an: vocal cord vibration (voicing), the place of articulation and the manner ofarticulation.

. As fu as voicing is concerned, mtmtion has been made in the

chapter describing the articulatory organs that the vocal cords

represent a key element in the articulation of speech sOlllldi If the

cords vibrate when we produce a sollnd, the sound thus uttered is

< Vocalic sounds are defined as sounds "prodnced ...nIt an oral caYity in wruch the most ramcal consnictimt does not OlCceed !bat flJllllli in the high 'Iowels [i} and [u) and with vocal cords fuat are positioned so as to allow spontmoous voicing; in producing llOIlvocalic ao,mils one or bofu of fuL"Se conditions me not s.dsfied." CQnsOlla.ntal so1.ll'l

J )

]

)

] ]

]

]

J

[J

[J

J

[J

'1

produced with a complete closure of the vocal tract fullowed by a sudden release ofthe air, the articulation is accompanied by a burst, a sort of explosioIL Such so).lIidS are consequ=tly called plostves. As fu" articulation involves a total Obstnicti01i (occlusion) ofthe'trac(ali alternative flame for such a coiJs01lll1lt is that of stop (Ro7l!~ (jcluzivii). It should be m=tioned, however, that the two. tenDs are not ,exactly synonymous, since~a.l.. s()unds are stops (the: air stream is blocked) but are not plosive smmils a:i 'their articUlatIon is not accompanied by an explosive burst. We must remember, fu=, tfurt t:1:W 'articulation of any plosive sound includes three distinct phases: dming the :first one, often called the approach, the articulators are moving together; preparing to plock the airstream; during the S!'COM stage, called the hold or closure, the articulators completely block the speech tract preventing the air to go out and contributing thus to builcting up the pressure of the airstream; during the third and last stage, called the release, or plosion, the speech organs move swiftly, releasing the air with an explosion.

If the stricture or narrowing of the tract does not result, however, in a complete blockage and 11 narrow passage is left for the air to go out, the pressure building up in the case of plosives is absent and the sound is uttered n~t with a sudderiburst, but continuously, the articuJation being accompanied by friction between the a:in;tream: and the speech organs. Such sounds'are therefure called fricatives. A major distinction between the latter and the stops is that they are conlinuani, noruibrupt sounds and their articulation can be, at least theoretically, continued mdefinitely.

The last major group of consonants that can be identified on

the basis of the triannet of articUlation is that of the affricates.

Affiicates combine the features of the two previou,s classes of sounds,

since their articulation starts like that of a ploSive, by a complete

. blockage of the airstream, but continues l.ilre that of a fricati~. as the ' ,.n"xt stage does not involve an abrupt release Of the air, but a gradual one. The symbols used in the phonetic transcription of these sooods in English are, as we shall see, somehow suggestive of their ambiguous, hybrid nature.

68

3.4. Sonorants. The Approximants: glides aJill] liquids

Not all continuant sounds are produced, however, with friction, as is the case of fricatives, mentioned above. There are sounds in English (and other IBlloouages as well, of course) the pronunciation of which dol'S not involve a major obstruction m the speech tract and does not produce the auditory effect ,of friction that characterize fricatives. Such sounds are co=only called approrirnants or frictionless continuants. The glides and the liquids are the two major

-! subclasses ofapproximants.

Tne glides are sounds such as [wJand fj] in English words like wife and young. Articulatorily,' they have a predDminimtly vocalic character since no maj or obstacle can be.identified when analyzing the way 'in which these sounds are utter~d. If this is a :feature that emphaSizes their vocalic character, their distribution is not, however, that ofa vowel; they can never be syllable nuclei (they are not syllabic in SPE terminology) and they always precede a genuine vowel. Because of their dual nature they are traditionally called semivowels or semiconsonants, the very coexistence of the two names suggesting the uncertainty and hesitation: of specialists, confronted with their ambiguous lllIfure. A more detailed description of glides will,be given later, when diphthongs are discussed.

Liquids constitute an important subclass of sonorants. Their high level of sonority places them, like the glides and the nasals, between vowels and genuine consonants. Liquids can be lateral sounds like [I] - the name comes from' the fact that when we utter these sounds the air is released laterally on one or both sides of the tongue - or rhona like [rJ - the Dame comes from the Greek word rho, ,designating 'the Jetter R in the Greek alphabet 1f in standard EngliSh'the sound'has the featureS of an approxllnant, more exactly of a glidf>.like sound, being produced without any kind of friction, in certaii:t dialeCts of'ErigJish when this sound is uttered the tongue is placed agaIDst the alveolar ridge and caused to vibrate, generating a

69

sonorous, intermittent sound a.q !he tongue touches the paSlrive articrtlator quickly and repeatedly, interrupting the outgoing airstream. It is the kind of [r] that appears in Spanish words like Rodrigo, real, etc. or in the interjection brrrrl that accompanies a shivering sensation.. It is called the rolled or trilled [rJ. If when the sowtd is uttered the tongue rapidly touches with only one movement the post-alveolar region we have a tap or flap type of thotic. If the tip of the tongue is drawn even further back" the thotic thus articulated is called retroflex.

3.5. Oml and nasal articulation

A different criterion that can he used to distinguish among sOlmds is !he position of ilie velum or soft palate (see. in Chapter 2 above the description of the main articulstory organs). Jfthe velnm is lowered, thus allowing the air 1:0 escape through the ll!!Slll cavity, we are dealing with a nasal sound. Ifit is raised, blocking the nasal eavity and letting the air out through the om! cavity (the mouth) the sound is called oral. English nasal con.qonants are stops as the airstream is caznplerely blocked when these cODSorumts are uttered, but they are not considered pIomve sounds as tqair release stage differs from. that of om! stops. Nasal sounds are son=ts and ofall the members of tlris class theydh'Play the lowest degree of sonority. As fur as English vowels are concerned, nasality is a contextual feature as we are going to see later.

3.6. Force ofarticulation

Another parnmerer differentiating among obstruents is farce of articulation. A greater llrtlculatory effort and a greater air pr!)Ssure required by a gre!rter J:esistance at the place of articulation (where the constriction takes place) cb~ sounds called fortis consonants while Ienis consonants are those obstruents the articula:ti.on of which requires a comparatively lesser effurt and a lower air pressUTe than .in

"'

the case oftheir fortis countexparts. The duration ofarticulation is also ] longer in the case of fortis sounds than in the case ofthe Lenis ones. In a voiced I voiceless pair, such as [t] I Cd], for instance, the feature fortis always characterizes the voiceless consonant, while the voiced [Joneill lani .

I Ifwe. consider the .mct iliat the source of the airstream that haq fJ an essential role in producing the sounds are the lungs, then we can I say that all the sounds of English are pulmonic (Latin puima, ,i pztlmollls, m.eaning Zuni). The direction of the airstream is from the [J; lungs out of the body, in other words we always speak during the ,I expiiation, not during the iospiiation phase of breathing. Therefore, ,

the sounds of Engli'lh are also egressive. There are, however, fIlanguages where the sounds are ingressive'as they are uttered while the speakers breathe in.

I 3.7.Place of articulation

i I I

'1 We have so far examined English consonants taking into I I account the manner iri which they are articulated. Another equally

!

:i:mpo.rtJmt criterion we can llSe in classitYing Engli'lh consonants is the

place where the obstruction is aclrieved, the plare of articula:ti.on. A I

,, tfurtinction has already been drawn between active and passive i articuJatars, which started from the comparatively higher or lower II degree of mobility of the organs Involved .in the articulati01l. As we are going to see, the :lk1Il1es giv

J

]

sounds (from the Latin word labium" meaning lip). If bofh lips are used to utte

3.8. The Description of English Consonants

l-Iaving examined the maID criteria we can use to classify consonants from M articulatory point of view, we can noW briefly describe the consonant phonemes ofEnglish.

A. The Approximants

1. The Glides. There are two SOlIDds in English, [w] and [j], baving vowel-like features as far as therr articulation is concerned, but which differ from their vowel counterparts [u] and [i] respectively through their distribution, force of articulation and length. WheIi we articulate a glide the articulatory organs start by producing a vowel-like souod, but then they immediately change their position to produce another souod. It is to the gliding that accompanies their articulation that these sounds owe their name. As we have seen earlier. precisely because of their ambiguous nature they are also called semivowels or semicoTlSonants. Unlike vowels. they cannot occur in syllable-final position, can never precede a consonant and are always followed by a genuine vocalic sound.

a. [w] is a labio-velar. rounded sound. At the beginning. its articulation is simi1arto that of the vowel [u]. but then the speech organs shift to a different position to utter a different vocalic sound. The distribution of 1he . sound includes syfu,),le-initiaI position before almost any engliSh vowel (e.g. win [wm], weed [wi.:d], wet [wet], wag [wcegj, work [wa:k], won [WN!],woo [wu:]. wood [wud], walk, [wD:k] warider [w~de],}oradiphthong (e,g.wczy). Before [r]. (e.g. write) the sound is no longer

. pronounced. [w] can also occur after a plosivc (e.g. twin, .. queen) or a fricative consonant (e.g. swine). It cim be

rendered graphically either by the letter w (the most common case) (e.g. sweet) orbyn (e.g. quite).

b. OJ is an unrounded palata). semivowel. The initial stage . of its pri>imnciation is quite similar to that of the short vowel [I], but then the 'sound glides to a different vocalic value. Like [w]; OJ cannot occur in final position (as a quite similar. palatal sound very often does in Romanian). is never followed by a consonant and occurs in front of back, central and front vowels. , (e.g. yes, young, youth). It ,can be preeeded by a plosive (e.g. tune) or a fricative (e.g.fome). The sound may be, spelt y (as in year) while -in words spelt with n. ne, m, ew, en and ean read as the long vowel [n:] the palatal sound is often inserted. Vte insertion is obligatory if. the preceding consonant is: an oral plosive (p, b, t, d, k, g), a nasal stop (m, n), a labio-dental fricative (f, v) or a glottal one (h). A word like beauty can only be read [bju:tr] and not [bu;tr]. Cf. also: pure, bureau, tUlip, deuce, .queue, argue, mule, neutrar. forious, reV1Ie, huge. The palatal 'solIDd is not inserted after affricates or after [r] or 0] preceded by a consonant: chtt', June, rude, clue. When [1] is not preceded by a consonant or when the sound preceding [u:] is an alveolar fricative [s, z] or a dental one, the usage varies: cf. suit [sjn:t], br also [sn:t]. In words like

.-

,.

. ! A rolled [rl i~ 'common in northern dialects and in,J i raises towards the hard palate. Words like lake [leikJ, look [Iuk], flute [flu:t], lurid [tjund] delight [dJIaIt]J illUb1:ra:te the duitributiOll f the consonant in syllable

;) initial position' or after a plosive plot [plot], ,Hake _[bIeik], clean [kIi:n] , glue [gIn:} or a fric.

I

and only exCeptionally gb in hiccough, The letter p is silent when followed by another obstruent or a nasal in word-initial position: psalm, pti!rodactyl, pneumatic.

b [bJ is the voiced, lenis counterpart of [p]. Voicing and, force of articulation are the 'f"mures that contra.'rt the two phonemes, (b J being like [p1a bilabial sound. It is distributed in all three basic positions; initial, medial and final: bet, above, cab. It is spelt b: about or bb:abbot The letter is silent in fum1. position