PACIFIC LINGUISTICS

S eJt..t e.6 V - No. 34

A GENERATIVE APPROACH TO THE PHONOLOGY

OF BAHASA INDONESIA

by

Hans Lapoliwa

(MATERIALS IN LANGUAGES OF INDONESIA, NO. 3)

W.A.L. Stokhof, Series Editor

Department of Linguistics

Research School of Pacific Studies

THE AUSTRALIAN NATIONAL UNIVERSITY

Lapoliwa, H. A generative approach to the phonology of Bahasa Indonesia.

D-34, vi + 160 pages. Pacific Linguistics, The Australian National University, 1981. DOI:10.15144/PL-D34.cover
©1981 Pacific Linguistics and/or the author(s). Online edition licensed 2015 CC BY-SA 4.0, with permission of PL. A sealang.net/CRCL initiative.
 P A C I F I C L I N G U I S T I CS is issued through the L�ng u��t�Q C��Q l e 0 6

Can b e��a and consists of four series :

S E R I ES A - O C C AS I O N A L PA P E RS

S E R I ES B - M O N O G RA PHS

S E R I ES C - B O OKS

S E R I ES V - S PE C I AL PUBL I C A T I O N S

EDITOR: S.A. Wurm.

ASSOCIATE EDITORS: D.C. Laycock, C.L. Voorhoeve, D.T. Tryon, T.E. Dutton.

EDITORIAL ADVISERS:
B. Bender, University of Hawaii K.A. MCElhanon, University of Texas
D. Bradley, University of Melbourne H. MCKaughan, University of Hawaii
A. Capell, University of Sydney P. Muhlhausler, Linacre College, Oxford
S. Elbert, University of Hawaii G.N. O'Grady, University of Victoria,
K. Franklin, Summer Institute of B.C.
Linguistics A.K. Pawley, University of Hawaii
W.W. Glover, Summer Institute of K. Pike, University of Michigan; Summer
Linguistics Institute of Linguistics
G. Grace, University of Hawaii E.C. Po lome, University of Texas
M.A.K. Halliday, University of Sydney G. Sankoff, Universite de Montreal
A. Healey, Summer Institute of W.A.L. Stokhof, National Center for
Linguistics Language Development, Jakarta;
L. Hercus, Australian National University University of Leiden
N.D. Liem, University of Hawaii J.W.M. verhaar, University of Indonesia,
J. Lynch, University of Papua New Guinea Jakarta

ALL CORRESPONDENCE concerning P A C I F I C L I N G U I S T I CS, including orders and

subscriptions, should be addressed to :
The Secretary ,
PAC I F I C L I N G U I ST I CS,

Department o f Lingu i stic s ,

Research School o f Pac i f i c S tudi e s,
The Austra lian Nationa l Univers ity,

Canber ra , A . C . T. 2600.
Austral ia .

Copyright 0 The Author

First published 1981

The editors are indebted to the Australian National University for

help in the production of this series .
This publication was made possible by an initial grant from the
Hunter Douglas Fund .

Na tional Library of Australia Card Number and I SBN 0 8 5883 2 4 5 3 .

 TA B L E O F C O NT E NTS

Page

ACKNOWLEDGEMENTS

CHAPTER 1 : INTRODUCTION 1
1.1. 'BAHASA INDONESIA' 1
1.2. THE AIM AND SCOPE OF THE STUDY 4
1.3. THE DATA: MATERIALS AND INFORMANTS 5
1.4. TRANSCRIPTIONS, NOTATIONAL CONVENTIONS, AND ABBREVIATIONS 7
1.4.1. Transcriptions 7
1.4.2. Notational Conventions 8
1.4.3. Abbreviations 8

CHAPTER 2 : THE RELATIONSHIP OF THE UNDERLYING REPRESENTATION

TO THE PHONETIC REALISATION 11
2.1. CONSONANTS 12
2.1.1. The Consonant System in Indigenous Words 14
2.1.1.1. The Underlying Consonant Ip l 15
2.1.1.2. The Underlying Consonant Ibl 16
2.1.1.3. The Underlying Consonant It I 16
2.1.1.4. The Underlying Consonant Id l 17
2.1.1.5. The Underlying Consonant Ikl 18
2.1.1.6. The Underlying Consonant 191 19
2.1.1.7. The Underlying Consonant lsi 19
2.1.1.8. The Underlying (Consonant) Glide Ihl 20
2.1.1.9. The Underlying Consonant IC! 21
2.1.1.10. The Underlying Consonant /JI 21
2.1.1.11. The Underlying Consonant Iml 22
2.1.1.12. The Underlying Consonant In l 22
2.1.1.13. The Underlying Consonant Ifll 23
2.1.1.14. The Underlying Consonant IQI 23

iii
iv

Page
2.1.1.15. The Underlying Consonant 111 24
2.1.1.16. The Underlying Consonant Irl 24
2.1.1.17. The Underlying Glides Iwl and Iyl 25
2.1.2. Loan Consonants 25
2.2. VOWELS 28
2.2.1. The Underlying Vowel /il 29
2.2.2. The Underlying Vowel lal 30
2.2.3. The Underlying Vowel luI 30
2.2.4. The Underlying Vowel 101 30
2.2.5. The Underlying Vowels leI and lal 31
2.3. THE CHARACTERISATION OF SEGMENTS IN TERMS OF DISTINCTIVE
FEATURES 35

CHAPTER 3 : POS ITIVE CONDITIONS 39

3.1. LEXICAL STEMS 40
3.2. AFFIXES 43
3.3. EXAMPLES OF UNDERLYING FORMS OF TYPICAL WORDS 46
3.4. A NOTE ON WORD ANALYSES 52

CHAPTER 4 : MORPHEME STRUCTURE CONDITIONS 57

4.1. SEQUENTIAL 'IF-THEN CONDITIONS' 59
4.1.1. Sequence of Consonants 59
4.1.2. Sequence of Vowels 65
4.1.3. Final Segments 72
4.2. SEGMENTAL 'IF-THEN CONDITIONS' 74
4.2.1. Consonants 74
4.2.2. Vowels and Glides 76
4.3. SUMMARY OF 'MORPHEME STRUCTURE CONDITIONS' (MSCs) 77

CHAPTER 5 : PHONOLOGICAL RULES 83

5.1. CONSONANT DEGEMINATION AND TRILL DELETION 84
5.1.1. Consonant Degemination 84
5.1.2. Trill Deletion 86
5.2. GLOTTAL STOP INSERTION, AND GLOTTAL STOP REALISATION OF Ikl 87
5.2.1. Glottal Stop Insertion 87
5.2.2. Glottal Stop Realisation of Ikl 88
5.3. DELETION OF Ih/, SYLLABLE REDUCTION, VOICING OF Ihl AND
PLACE ADJUSTMENT OF Ihl 90
5.3.1. Deletion of Ihl 90
5.3.2. Syllable Reduction 93
5.3.3. Voicing of Ihl 96
 v

Page
5.3.4. Place Adjustment of Ihl 97
5.4. NATURALISATION RULES 98
5.4.1. Naturalisation of IfI 98
5.4.2. Naturalisation of Izl 100
5.4.3. Naturalisation of lsi 101
5.4.4. Naturalisation of I xl 102
5.5. NASAL ASSIMILATION, SCHWA INSERTION AND CONSONANT DELETION
(COALESCENCE) 103
5.5.1. Iman l Before Stems 104
5.5.2. I ma n l Int�rmediate Between Ipal and Stems 112
5.6. VOWEL: LAXING, RETRACTION, DELETION, NASALISATION; LOWERING
AND COALESCENCE 114
5.6.1. Vowel Laxing 114
5.6.2. Vowel Retraction 115
5.6.3. Schwa Deletion 116
,
5.6.4. Vowel Nasalisation 118
5.6.5. Vowel Lowering 119
5.6.6. Vowel Coalescence 120
5.7. GLIDE INSERTION AND DESYLLABICATION 121
5.7.1. Glide Insertion 121
5.7.2. Desyllabication 122
5.8. VOWEL DISSIMILATION 123
5.9. LOANWORD ENDINGS 126
5.10 . STRESS PLACEMENT 127
5.11. SUMMARY OF P-RULES 131

CHAPTER 6 : SUMMARY AND CONCLUSIONS 143

BIBLIOGRAPHY 147
Lapoliwa, H. A generative approach to the phonology of Bahasa Indonesia.
D-34, vi + 160 pages. Pacific Linguistics, The Australian National University, 1981. DOI:10.15144/PL-D34.cover
©1981 Pacific Linguistics and/or the author(s). Online edition licensed 2015 CC BY-SA 4.0, with permission of PL. A sealang.net/CRCL initiative.
 CRAPTER 1

INTRODUCTION

T h i s study i s a trans format ional generat ive a c c ount of the p hono l o g y

of ' Bahasa I ndone s i a ' , t he nat ional and o f f i c i a l language o f t he Repub­
l ic o f I ndone s i a . Eac h o f t he subsequent c hapt er s b eg i n s w i t h int r o ­
duc t or y r emarks on a number o f p o int s , su c h a s t heoret i c al i s sue s o n
c ert ain a spe c t s o f phonolog y , t he m o d e l o f d e s c r ip t io n adop t ed , t he
ob j e c t ive and orga ni sat i o n of t he c hapt er c oncerned , e t c . Ther e f or e
t h i s i nt roduc t or y c hapt er pr e sent s o n l y a brie f a c c ount o f eac h o f t he
f o l l owing :
(a) ' Baha s a Indone sia '
( b ) t he aim and s c o p e o f t he study
( c ) t he data - mat e r i a l s and informant s
( d ) t ran s c ript i on s , not at ional c onv ent i o n s , and abb reviat i o n s

1.1. 'BA H A S A I ND O N E S I A'

l
The t erm ' Bahasa Indone s ia ' meaning ' Indo n e s i an language, was first
introduc ed b y the S e c on d All I nd o n e s i a You t h Congr e s s in J akart a . In
t he ir appeal for t he unit y o f var i o us regional you t h organ i sat i on s , s o
a s t o make t heir independen c e movement str o nger and mor e e ff e c t ive , t he
d e l egat e s t o t he c ongr e s s p a s s e d a r e s o l ut ion on 28 O c t o b er 1 92 8 . Thi s
r e s o l ut ion , known a s Sumpah Pemuda ( Yout h P ledge ) c alled fo r l o ya l t y t o
one fat her land nam e l y ' Fat her l and I ndone s i a ' , t o one nat ion n am e l y
' Ind o ne s ia ' a n d t o o n e language nam e l y ' Bahasa I n done s i a ' . B y virt ue
of t hi s r e s o l ut io n , M a l a y , t he language o f east ern Sumat r a and t he
n e arb y i s land s ( and o f t he Mal a y p e n i n s u l a o f Malaysia as we l l ) , was
adop t ed and c hr i s t ened ' Bahasa Indone s ia ' ( hereinafter abbreviated B I ) ,
and wa s c on c o m i t ant l y promo t e d t o nat ional l a nguage s t at u s . The

l
The word baha sa itself i s derived from Sanskrit b hasa ' la nguage ' .

Lapoliwa, H. A generative approach to the phonology of Bahasa Indonesia.

D-34, vi + 160 pages. Pacific Linguistics, The Australian National University, 1981. DOI:10.15144/PL-D34.1
©1981 Pacific Linguistics and/or the author(s). Online edition licensed 2015 CC BY-SA 4.0, with permission of PL. A sealang.net/CRCL initiative.
2

adopt io n o f Malay as t he nat ional language of I ndone s i a under t he new

name caused no pro b l em mainly b e c au s e Malay had been used a s a l ingua
franca t hroughout the archipe lago long b e fore t he c o l on i a l p e r i o d b egan
in t he e ar l y year s of t he 1 7 t h century . Moreover , t here was no ot her
regional language c ompet ing for t he nat ional l anguage stat us . As a
mat t er o f fact , s inc e O c t ober 1 9 2 8 B I ha s served as a p ower ful l everage
for nat ional unit y and as a main mean s o f int er-et hnic and int e r ­
c u l t ural c o rr�un i c at ion in Indone s i a .
Soon after Ind on e s i a pro c l aimed her ind epend enc e on 1 7 Augu st 1 9 4 5 ,
B I was c onst itut ionally made t he o f f i c i a l language o f t he Repub l i c o f
1
I ndone s i a . Then c e .B I ha s b e c ome a language b y means o f whi c h maj or
government a l a f fa i r s ar e c arried out , a l a nguage o f instru c t ion at a l l
l eve l s o f I ndones ian s c hoo l s , e x c ept , o f c our s e , in fore ign l anguage
c la s s e s at sec ondary s c hoo l s and univer s i t ie s , a s w e l l a s in foreign
language depar tment s of unive r s it i e s or c o l l ege s where t he fore ign lan­
guage t a ught i s used inst e a d ( c f . H a l im 1 9 7 4 : 8) . H o wever , in regions
where B I i s c o n s iderab l y d i f ferent from t he l o c a l language s , B I is not
used a s t he medium of i n s t r uc t i o n unt i l t he t hird or t he four t h grade
of e l ementary s c hool s .
I t must be n o t e d however t hat for mo st speaker s , B I i s t he i r s e cond
language . I ndone s i a i s a mult i-et hn i c and mult i - c u lt ura l count ry .
2
I t s p opu lat i o n , e s t imat ed t o number over 1 3 0 m i l l i on p e o p l e b y t he end
of 1 9 7 7 c on s i s t s of hundreds of e t hn i c group s , and subsequent ly a lmost
eac h one of t hem has it s own l anguage or dia l ec t . A c c ording t o some
s c ho l ar s , t here are about 4 0 0 language s ( in c luding d ia le c t s ) t hroughout
t he c ount ry ( c f . H a l im 1 9 7 5 ; Ro s i di 1 9 7 5 , et c . ) . Genet i ca l l y , t he s e
l anguages ( in c l ud i ng Malay , and t hu s B I ) b e l ong t o the Indone s i an group ,
and t oget her w i t h t he languag e s o f Madagas c ar , Formo sa and t he
3
P hi l ip p in e s con st i t u t e t he Indone s ian sub - fami ly o f what i s c urrent l y

l
It i s stated in the Undang-Undang Va6� 1945 ( 1945 Const itut ion ) Chapter XV, Art icle
36 that "Bahasa Negara adalah Bahasa Indonesia" ( The State language is Bahasa
Indone sia ) .
2
�'he figure is a rough est imate made on the bas i s of approximately 2 per cent popu­
la.t ion rate per year of 120 . 1 million of the 1971 population c ensus ( c f. Iskandar
1 975 : 5 ) .
3
Since the term ' Indonesian ' has been used to refer to one of the sub-families of the
Austronesian family as well as to the languages of the Indonesian archipelago , it i s
desirable to use the term B I to refer exclusively t o the national and o ffic ial lan­
guage of Indonesia, and let the t erm Malay refer to the language of eastern Sumatra
ani the nearby i slands only ( cf . Halim 1974 : 9ff) .
 3

l
known as t he Austrones ian fam i l y . Apart from M a l a y , whic h i s t he
b a s i s o f B I , Javane s e anq Sundane s e need ment ion in r e lat ion t o t he
devel opment o f B I .
Javane se , t he language s p o ken in c entral and east J ava by about h a l f
o f t he populat ion o f t he who l e arc hipe l ago , ha s prob a b l y b e e n t he most
in fluential loc a l language in t he devel opment o f B I , p ar t l y b e c au s e
t he maj o r i t y o f t he nat ional l e ader s a r e speaker s o f J avan e s e a n d p art l y
b e c au s e man y wor k s i n J avan e s e o r about Javane s e ( in c luding it s l i t e r ­
ature a n d c u l t u re ) have gained w i d e r e a der s h ip b o t h i n t he c ount r y and
abroad . The sec ond m o st infl uent i al l o c a l language in t he development
of B I , for s imi lar r e a s o n s ment ioned above , is probab l y Sundane s e , a
language spoken by t he maj or i t y o f t he nat ive inhab i t a nt s o f we st Java .
It mu st b e n o t e d however t hat t he impor t a n c e o f t he regional langua g e s
in t he devel opment o f B I i s l im i t e d t o t he enr i c hment o f it s v o c ab u lar y .
Gr amma r and pronun c iat ion i n f l uenc e s are unac c ep t a b l e i n t he s e n s e t ha t
t he y a r e r egarded a s bad usage . A s we s h a l l s e e lat er in Chap t er 4 ,
t he imp a c t o f t he influence o f t he r e g i o na l language s ( e s p e c ia l l y
Javane s e and S undane s e ) o n t he phono logy o f B I i s marked b y t he e x i s ­
t en c e o f a number o f morpheme st ructures i n B I w hi c h are , p honot a c t i c ­
a l l y , unc ommon wit h r e sp e c t t o t he structures o f Mal a y it ems , s u c h a s
t he oc c urr e n c e o f sc hwa i n the f i n a l s yl lab l e as in g a n t e n g [ g a n t a Q ]
' hand s om e ' , t he oc c urre n c e o f a front m i d vowe l i n t he penult ima t e
s y l l a b l e i n wor d s who s e f i n a l s yl l a b l e c on t a i n s a h igh vowel as i n
b e n d i [ b e n d i ] ' a gig ' , etc . T h e ex i s t en c e o f t ho se ' unc ommon ' mor p heme
s t ru c t ur e s i s part i a l l y at t r ibut ab l e t o t he i n f l ue n c e o f foreign lan­
guage s .
Among foreign l anguage s whic h p l ay a s i gn i f i c ant r o l e i n t he d e v e l ­
opment o f B I , part i cu l ar l y i n t he enr ic hment o f i t s v o c ab ul ar y , A r ab i c ,
Dut c h and Eng l i sh merit ment ion . Arab i c has a remarkab l e i n f l u e n c e
ma i n l y t hrough t he M o s l em r e l igion , t he r e l i g i on o f the v a s t maj o ri t y
o f t he p e o p l e , wh i l e Dut c h and Eng l i s h d i sp lay t h e i r i n f l u e n c e t hr ough
mod e rn s c i en c e s , t ec hno lo g y and c u lture as we l l a s t hr ough c ont a c t s
wi t h t he we st ern wor l d i t se l f . T he impact o f s u c h i n fluen c e on t he
phono logy o f BI i s marke d , in ad dit i o n t o t he e x i st en c e o f t he ' un c ommo n '
morp heme s t ruc t u r e s ment ioned abov e , b y the ac c e p t an c e · or r e c o gn i t i on
o f some s p e e c h s ounds and sound c omb i nat i o n s whi c h e x i st n e i t he r i n
Mal a y n o r i n t he r e g i o n a l language s , such a s t he fr i c at ives / f / ( e . g .

l
Traditionally the Indonesian languages are grouped under the Malayo-Polynesian family,
a term introduc ed by K.W. von Humboldt , who owed i t , to some extent , to Forster ( see
Fraser 1892 : 344-345 ) . For a further account of the term 'Austronesian', the r eader
is referred to Brandstetter 1916 : v , Halim 1974 : 9 and the referenc es therein .
4

in f a h am ' c ompr e hend, i de o l o g y ' , v e t o ' v e to ' , et c . ) , Izl ( e . g . in z a m a n

' era ' , zat ' s ub s ta n c e ' , et c . ) , lsi ( e . g . in syah ' l ega l ' , syarat 're­
qui r e m en t s ' , e t c . ) and Ixl ( e . g . i n k h a s ' ty p i c a l ' , khu sus ' sp e c i fi c ' ,
et c . ) , c o n sonant c lust er s in word -init i a l p o s it ion a s in p r a k t e k
' p ra c t ic e ' , struktur ' s tr uc t ure ' , e t c . ( See furt her d i s c u s s i o n o f
t he se p henomena i n Se c t i on s 2 . 2 . , 3. 1. and 5 . 4 . For t he c orre spondenc e s
b e t we e n phonol o g i c a l s ymbo l s and ort hograph i c symbo l s , s e e s e c t i o n 1 . 4 . )
Ob servat ion o f t he sp e e c h form o f B 1 shows t hat the language i s far
from uniform . Thi s can e a s i l y b e under s t o o d b e c au s e t he g eograp hi c a l
and l ingui st i c bac kground o f t he speaker s i s f a r from homogeneou s . The
t yp e of BI d e s c r i b e d in t h i s s t ud y is of what may b e t ermed ' St and ard
Pronun c iat i on ' , a kind of s p e e c h form ( pronunc iat i o n ) whi c h is u sual l y
u s e d b y educ at e d p e o p l e a n d c on s idered appropr iate t o u s e i n p l a c e s
l i ke schoo l s , o n radio and t e l e v i s i on , i n o f f i c e s , et c . The ' St andard
l
Pr onun c i at ion ' var iet y is geograph i c a l l y least marked .

1.2. THE A I M A N D S CO P E O F THE STUDY

The a i m o f t hi s s t u d y is t wofold . One goal is t o pre sent an int r o ­

duc t ion t o t he p honologi c al grammar o f B I . I t i s an introduc t ion
b e c au s e it doe s not ac count for all p hono l o g i c a l a s p e c t s o f BI and it
is l imit ed t o one t ype of sp e e c h var i e t y of B I . It is a l s o an intro ­
duc t i o n in t he s e n s e t hat t hi s study i s an a t t empt t o provide a gener­
2
ative a c c ount of t he p ho no l og y o f B I .
T he ot her - pr imary - goal i s t o pr e se nt a p hono l o g i c a l d e s c r i p t i o n
o f B I w it hin a t ran s format ional generat ive framework . M o r e spe c if i c a l l y ,
t hi s st udy i s c on c erned w i t h t hree main prob lem s , name l y : ( 1 ) t he r e p ­
re sentat ion o f B I morp heme s b o t h on t he phonol o g i c a l and phonet i c level ;
( 2 ) t he s t at ement of t he we ll -formedn e s s o f BI morpheme s and s e gment s ;
and ( 3 ) t he formulat i o n o f r u l e s whic h c orrect l y acc ount for t he p hone t ic
form (in t hi s c a se ' St andard Pronunc i at ion ' ) o f a l l ut t eranc e s in B I
and at t he same t ime adequat e l y capt ure p hono logical genera l i s at i o n s
o f t h e l a ng uag e .
The framework adopt e d in t he present de script i o n i s b a s i c al l y t ha t
o f C homsky a n d Ha l l e as d e s c r i b ed in t he i r monumenta l work T h e S o u n d
Pa..t.teJtn 0 6 Eng.ii.6h ( 1 9 68) . I n d e a l ing w i t h t he prob lem o f t he we l l ­
formedne s s o f mor p hem e s however , t he mod el used depar t s from C homsky

l
Ideally 'Standard Pronunciation' i s geographically unmarked ( c f . Abercrombie 1956:
44 ) .
2
Halim ( 1974:7) points out that the least explored part of BI grammar is its phonology .
Thi s is part icularly true with respect to the use of a generative approach .
 5

and Hall e ' s in t hat regular i t i e s ob served in B I morp heme s and s egment s
are , fo l lowing Stanl e y ( 1 9 6 7 ) , s t at ed in t erms of c ondi t i on s rat her
t han i n t erm s of r u l e s ( c f . Chom s k y and Hal l e 1 9 6 8:1 7 1 ) .
The sc ope of the pr e s ent de s c r i p t i o n i s l imit e d t o t he word - l e v e l
p hon o l o g y . I nt onat i o n a n d s ynt ac t i c a l l y-mot ivat ed s t re s s a r e not d i s ­
c u s sed . T o deal wit h t he se a spe c t s would c er t a i n l y r e s ult i n ano t he r
study.
T h i s st ud y cons i s t s o f s i x c hap t er s i n c l uding the Introduc t i o n .
C hapt er 2 i s devot e d t o t he d i s c u s s ion o f segment s and fe at ur e s whi c h
are nec e s s ar y for t he r epre sent at ion o f morphemes . Chap t er 3 deal s
wi t h t he st at ement of t he wel l - formedne s s o f l e x i c a l and grammat i c a l
morpheme s a s w e l l a s morpheme c ombinat i on s in t erms o f p o s i t ive c ond i ­
t ions . The p o s i t ive c o nd it i o n s app l y o n l y t o t he p hono l o g i c a l ( under­
l ying ) form of morpheme s in B I . In Chap t e r 4 t he que st i o n of t he mor ­
p heme we l l - formedn e s s i s t aken up aga i n . Whi l e Chapt er 3 i s c onc erned
mai n l y w it h t he c anoni c a l s hap e s of morpheme s o n t he phono l o g i c al l e v e l ,
Chapter 4 i s princ i p al l y conc erned wit h p ho no l o g i c a l regular i t i e s ( or
p hono l o g i c a l r edundan c i e s ) o b served in t he underl ying repre sent at i on
of morp hem e s and s e gment s . The ob se rved regular i t i e s are s t at e d i n
t erms o f ' I f-Then ' c ondit i on s . The s e condi t i o n s are c a l l e d ' Morpheme
St ructure ' c o n dit i o n s . Chapt er 5 i s conc erned wi t h t he formulat ion of
p honol ogic al rul e s proper ( a l s o c a l l e d p honet ic real i s at i o n r u l e s ) .
The main func t io n of t he s e rule s i s t o a c c ount for v ar iat i o n s i n shape
t hat morphem e s undergo in var i o u s environment s . F in al l y , Chap t er 6
pr e sent s t he summary of t he s t ud y and a number o f c onc l ud i ng remarks
drawn from t he ana l ys e s made i n t he ear l ier c hapt er s .

1.3. THE DATA: MATE R I A L S A N D I N F OR M A N TS

A s not e d above ( c f . S e c t i o n 1.1.), t he var i e t y of B I de s c r ib e d in

t hi s study is t hat o f ' St andard Pronunc iat ion ' which i s d e fi n e d a s a
kind o f spe e c h form of B I whi c h i s usual l y u s ed b y educ a t e d speake r s
and ge ogra p h i c al l y i s l e a st marked . Alt hough t he wr i t er i s a n ' e ducat e d
speaker ' o f B I , h i s pronun c i at ion i s not b y any mean s free from t he
infl uenc e o f hi s fir st language ( i . e . Mor i , one o f the language s spoken
in C en t r a l Sulawe s i } . To m i n im i s e the reg ional fe a t ur e s , t he wr i t er
has had r e c our se t o t he s p e e c h of ot her educat ed speaker s . For t h i s
purpo s e , t wo s e t s of mat e r i al s are u s e d .
The first s e t , t he p r imar y dat a , i s t he r e c or d i ng s o f 1 8 0 words
( i n c l uding p hra s e s ) , 50 s imple s e nt e n c e s and a dial ogue of 40 s e nt enc e s
l ong . The r e c or d i ng s wer e made i n t he p hone t i c labor at ory o f t h e
Phonet ic s a n d L i ngui s t ic s Department , S choo l o f Or i e n t a l a n d A fr i c an
6

S t u d i e s , Univer s it y of London . To avoid unn e c e s sary mi st ake s or d i s ­

t ort i o n s , t he t e x t o f t he language materia l s w a s given t o every i n form­
ant a few d a y s b efore t he r e cord ing o f h i s or her reading was c arried
out .
The informant s , f ive p e o p l e a l toget her ( inc l ud i ng t he wr i t er ) , are
Indone s i ans in t he ir 3 0 s and 4 0 s . T h ree of t hem , a l l men , are p o s t ­
gradua t e student s a t t h e Univer s i t y o f Lond on . They c ome from d i f fer­
ent p art s of I ndone s i a - Cent r a l Java, Nor t h Suma t ra and Cent ral
S u l awe s i . The o t her t wo , hu sb and and wif e , are emp loyed by t h e uni ­
versit y . B o t h o f t hem c ome from We st S umat ra .
In a dd i t ion t o t he a b ove mat erial s , t he wri t er a l s o u s e d anot her s e t
o f mat erial s ( s e c ondary dat a ) a s a furt her c he c k on h i s pronun c i at ion .
The se mat er ial s c on s i st o f ab out 1 5 0 wor d s ( in c luding phra s e s ) , 2 5 0
l
s e nt e n c e s and a conversat i o n o f 4 0 sent enc e s long . The s e s e c ondary
mat e r ia l s are on a ready-made t ap e ava i l a b l e in t he language laboratory
of t he S chool of Orient al and A f r i c an St udie s , Univer s i t y o f London .
I n forma t ion about t he spe aker s invo lved in t he rec ording s i s no t ava i l ­
able . As far as can b e j udged from t he r ec o r d i ng s , t he r e are four o f
t hem , t wo ladie s and t wo men .
The st at ement s about t he f e a t ur e s o f B I ' St andard Pronun c i at i o n ' are
der ived mainly from t he pr imary dat a on t he b a s i s o f t he feat ure s c ommon
t o t he informant s . The s ec ondary dat a are o c c a s i ona l l y consult e d when
the wr it er is i n doubt a b out a c er t a i n a sp ec t . A lt hough t he st at ement s
are b a s ed on t he anal y s i s o f a body o f ut t er an c e s ( corpus ) , t h i s s t ud y
i s b y no mean s a corpus-re s t r i c t e d de s c r i p t i on o f B I phonol ogy . Many
it ems u s e d to i l lust rat e c er t ain p o i nt s in t he c o u r s e of t he study are
sup p l ied b y t he wr i t er e it her from h i s own knowl edge of t he language or
from o t he r sour c e s such a s d i c t ionari e s , grammar b o o k s , e t c . , e spe c ia l l y
in d e a l ing w i t h t he prob l em s o f t he rep r e s ent at ion and w e l l - formedne s s
o f morpheme s . Thi s p r o c edure i s j u st i f i ab l e b ec au s e t he main d i fference
bet ween B I ' St andard Pronunc i a t i o n ' and t he regional var i e t i e s of t he
l anguage l i e s , as far as generat ive p honol ogy i s c o n c e rned , in t he
s y s t em o f p hone t i c rea l i s at io n rul e s . Cry s t al ( 1 9 6 9 : 8 ) wr it e s

T o g o b e y o n d t he c or p u s i s e nv i s ag e d from t h e v e r y o ut s e t , a s
u l t ima t e ly o n e w an t s t o m a k e s t a t e m e n t s ab out the l a n g ua g e
s y st em a s a w h o l e - or , t o put i t i n c u r r e n t g e n e r at i v e t er m s ,
a b o ut t he u n d e r l y i ng 'c omp e t e n c e ' wh i c h l i n gu i s t i c p e r f o r m a nc e
i s s up p o s e d t o r ef l e c t . . .

l
The materials were originally prepared for foreigner s learning BI ( cf . Johns 1975:
Chapter 13 ) .
 7

1 . 4. T R A N S CR I P T I O N S , N OTAT I O N A L CO N V E N TI O N S , A N D ABBR E V I AT I O N S

1 . 4 .1. T R A NS CR I PT I ONS

I n t h i s study, t here are t hr e e kind s o f t r an s c ript ions u s e d for t he

r e pr e s ent at ion o f B I i t e m s - ort hograp h i c , p honolog i c a l ( s ys t emat i c
p ho n emic ) a n d p ho net ic t r a n s c ript ions . A sample B I it em m a y b e g iven
in ort hographic and/or p hono logi cal , and/or p honet ic form . Ort hograp h i c
forms o f B I i t e m s a r e in b o l d p r int . Phon o l o g i c a l and p ho ne t i c forms
are wr i t t e n between s l a s he s ( / /) and square b r a c ke t s ( [ J ) r e sp e c ­
t iv e l y .
The r e lat i o n s h i p s b et ween phono l og i c a l forms and t he c orre sponding
p ho net i c form s are d e s c r i b e d in Chapt er 2 . The r e l at ionshi p s b e t ween
phono l o g i c a l forms and t he c orre spond i ng ort hograp h i c forms c an be
d e s c r i b e d b y s howing t he c orre sp ondenc e s between ( s ys t e mat ic ) p honeme s
and graphem e s ( i . e . minimal d i st inct ive unit s o f t he ort hography) in
BI as f o l l o w s :

Systematic Phoneme Grapheme Examp le

Iii l i bul i bu ' mo t her '
lui u I sa t ul - satu ' one '
I eI ==--=--=--==--- e lenakl - enak ' de �i a i o u s '
lal l am a s l - emas ' go �d '
101 0 I t o ko l - toko ' s ho p '
lal a lapal - apa ' wha t '
Iyl y layahl - ayah ' fa t her '
I wl w I ka wa n l k a wa n ' fri e n d '

�
-

Ihl h Iha r i l - ha r i ' da y '

I P/ p l apal - apa ' w ha t '
b / J a wa p l - j a wa b ' a n swer '
Ibl
�: I b uk aI buka ' to o p e n '
-

ItI I t u t upl - t u t up ' to a �o s e '

labatl a b ad ' a en t ur y '

�:
-

Idl I d u d u kl - duduk ' to s i t '

I k/ I k ua t l - kua t ' s trong '
I g u d e k l - g u deg , spiaed riae '
Igl Ig u l a l - gu l a ' s ug a r '
I fI -=====-==== f, p Ita ra fl - t a r a f , t a ra p ' p ha s e '
v I f e tol - veto ' veto '
l si s Isayal - saya 'I'

I zl z,s lazasl - a z a s ,a s a s ' p r i n a i p �e '

-=::::: j I za m a n l - j am a n , zama n ' era '
lsi sy , s I �ahl - s ya h , s a h ' �e g a � '
8

Systematic Phoneme Grapheme Example

I x/ � kh I xa s l -
khas ' ty p i ca l '
g l a x e nl - agen , a g e n t , a g e n cy '
I C/ c I cu c i l - cuc i ' t o wa s h '
!j l j laJa rl -
ajar , t o teach '
I II 1 / la r i l - 1ari ' t o run '
I rl r I r um a h l - r u m a h , h ouse '
Iml m Imud a l - muda ' y oung '
Inl n linil - ini ' this '
Ipl ny I b ap a k l - b a n y a k ' ma n y , m u c h '
I rJI ng l o r a rJI - o r a ng ' pe rs on '

1.4.2 . NOTA T I O NA L C O N V E NT I ONS

For re ferenc e , l i s t s o f samp l e i t em s in e a c h chap t er are numb e r e d

c o n s e c ut iv e l y b e g i nn ing wit h t he number o f t he chap t er , s im i l ar t o t ho s e
o f t he s e c t io n s o f t he c hapt er . Numb e r s re ferring t o e x amp l e s are
d i f ferent from numb e r s re ferring to s e c t io n s in t hat t he y are a lways
enc l o s e d bet ween brac ket s .
The f o l lowing are not at ional c onvent ions u s e d in t h i s s t ud y :

[J ' square brac ket s ' e nc l o s e p honet i c i t ems

/ / ' s l a s he s , s lant l i ne s ' enc l o s e phono l ogic al it em s
o p hr a s e boundary
+ grammat i c a l morpheme boundary
++ l ex i c al morp heme boundar y , word boundary
+ ' rewr i t t e n a s '
+ ' derived from ' , ' anal ysed i nt o '
� ' agree wit h ' , ' must b e ' , ' go t o ge t her w i t h '
* ' re c o n s truc t e d or h yp ot het i c al f o rm ' , ' unacc eptab l e form '
o ' z ero '

1 . 4.3 . A B B R E V I A T I O NS

Ant Ant erior

BI ' Bahasa I ndon e s ia ' ( ' Ind on e si a n ( l anguage ) ' )
C C o n s o nant
Cons Con sonant al
C ont C ont i nuant
C or Coronal
I PA I nt ernat ional Phonet i c Alphab et
MSC Morp heme Structure Condit ion
NP Noun Phrase
 9

P-rul e ' Phono l o g i c a l rule - p roper '

Son Sonorant
Syll Syllabic
V Vowe l
VP Verb Phr a s e
 CHAPTER 2

THE RELATIONSHIP OF THE UNDERLYING REPRESENTATION

TO THE PHONETIC REALISATION

A c c ording t o H arms ( 1 9 68 : 1 2 ) ,

T h e p r im a r y aim s o f g e ne r at ive p h o n o l o g y are t o pr o v ide a

p h o n emic r e pr e s e nt at ion o f m o r pheme s a n d a s e r i e s of o r d e r e d
r u l e s t hat , t o g et he r wit h i n f o r m at io n a b out b ou n d a r y p h e n o m e n a
( j u n c t ur e s ) , ( l ) a d e quat e l y e x pr e s s t h e p h o n o l o g i c a l g e n e r ­
a l izat i o n s o f t he l a nguag e a n d ( 2 ) at t h e s am e t im e d e t e r min e
t he p h o n e t ic f or m o f a l l ut t er anc e s in t he l a nguag e .

I n t he l it erat ure , t he t erm underlying representation ( or phono l o g i c a l

r e pr e s e nt at ion ) i s more frequent l y u s e d b y generat ive phono logy wr i t er s
1
t o mean pho nemic r e pr e s ent at i o n . S im i l ar l y , t h e t erm under lying ( o r
phono l o g i c al ) segment i s pre ferred rather t han t he t erm phoneme .
The not i o n o f underlying (phonologica l ) form o f a morpheme has b e e n
a s so c i at ed w i t h ' ab s t r a c t ne s s ' and t h i s ha s b e en t aken most l y t o mean
t hat t he unde r l y ing form r e pre sent s one or more phonet ic forms of t he
morpheme in quest i o n . I n pract i c e , t he und e r l y ing form c ho sen for a
c er t a i n morpheme i s u s ua l l y ident i c a l wit h one of t he o c c u rring phonet i c
2
forms o f t he morpheme . Thi s ho wever i s not a lway s t he c a s e . The
l i ng u i st may post u l a t e an und er l ying form for a morpheme whic h doe s not
o c c ur phone t i c a l l y in any environment . Yet t he po st u la t e d under l y i ng
form u su a l l y c on s i st s of e lement s ( s egm e nt s ) wh i c h are phonet i c a l l y

l
In fact the term 'underlying ( phonological ) representation' in generative phonology
is similar to the term 'morphophonemic representation' rather t han 'phonemic represen­
tation' in structural linguistic s . This term (morphophonemic ) i s avoided however
because it implie s t hat there is a phonemic level of representation between phono­
logical ( lexic al ) and phonetic representation ( cf . Chomsky and Halle 1968 : 11 ; Chomsky
1964 : 69 ; 1966:76-81 , et c . ) .
2
This is the main reason why many generative phonologist s have the view that the
phonological forms of morphemes must be fUlly specified in t erms of phonetic features
( e . g . Stanley 1967 ; Ander son 1974 , especially Chapter 3, etc . ) . This view is adopte d
in this study .

11
12

po s s i b l e in t he l anguage . T h i s is t o say t ha t t he under l y i ng s egment s

po s t ulated for t he unde r l y ing repre s e nt a t i o n are not s imply a b s t ra c t
segment s in an ar b i t rary formula but rat her , t h e y a r e i nt ended t o hav e
an e xpl ic it pho ne t i c int erpretat ion ( c f . Ander son 1 9 7 4 : 4 7 ) .
I n t hi s c hapt er we are go ing t o examine and det ermine what segment s
are n e c e s s ary for t he underlying repr e sent at ion of B I morpheme s and
t he i r phonet ic r e a l i s at ion . S ec t ions 2 . 1 . and 2 . 2 . deal with c o n s o nant s
and vow e l s r e s pec t iv e l y . S e c t ion 2 . 3 . pre sent s t he featur e c ompo s it ion
o f t he segment s set up in t he f i r s t t wo se c t i on s .

2. 1. CON S O N A N TS

M o s t works on t he pho n o l ogy o f B I rec ogn i s e t hat BI makes u s e o f

2 3 c o n sonant sound s . The s e c onsonant sounds are u s u a l l y r e pre sent e d b y
t he symb o l s : [p, b, t, d, k , g, 7, f, 5, Z, $, x, h, c, J , m, n, )1, I) ,
1, r, wJ and [yJ . ( The symbo ls [5], [c J , [J J and [yJ are equivalent t o
I P A symb o l s [9 J , [t9 J , [d� J and [jJ re spe c t ive l y . ) The s e c o n sonant s
may furt her b e c harac t e r i s ed in gener a l phonet i c t erms by mean s o f the
fo l lo wing c hart :

CHART 1

Place of Artic ulation

Manner of
Art iculat ion �ntal/ ( Alveo-)
Labial Vel ar Glottal
Alveolar Palatal

1. Plosives ( vl ) p t k 7
( Vd) b d g
2 . Fricatives ( vl ) (f ) 5 m (x) h
(vd) ( z)
3 . Affricates ( vl ) c
(vd) J
4 . Nasals ( vd) m n )1 I)
5 . lateral (vd) 1
6. Trill (vd) r

7 . Approx:!mants (vd) w y
( sEmivowels)

( Note : vl = vo:1celess; vd = voiced)

Not e t hat t he symb o l s e n c l o s e d b e t ween b r ac ket s in t he above c hart do

not oc c ur in nat ive B I ( Malay ) it ems . They have b e en int r o duc e d int o
the s o und s y s t em o f B I mainly t hr ough Arab i c bo rrowing .
 13

T he maj or i t y of t he works on t he phonology o f B I ment i o n e d above

c laim t hat t he 23 c o n s o nant sound s are lingu i st i c a l l y s ignifi c ant in
t he s e n s e t ha t t he y are c apab l e o f d i fferent iat ing t he meanings of
words and , ac c ordingly, e a c h o f t h em qua l i fie s for t he l ab e l phoneme
( s ee e . g . H a l im 1 9 7 4 : 1 7 5 - 1 7 6 ; a l s o H a l im 1 9 7 2 : 1 5 3 - 1 5 4 ; Wo l f f 1 9 7 1:9 ;
M a c donald and Dar dj ow i dj o j o 1 9 6 7 : 1- 2 1 , et c . ) .
Some s c ho la r s however po s it 2 2 c o n so nant phon eme s o n l y i n s t e ad o f
23. The y c l aim t hat t he glo t t a l s t op ( [?J) do e s n o t c on st it u t e a pho ­
neme but rat he r it i s o n l y an a l l ophone of t he phoneme /k/, who s e
d i st ribut i on i s phono logi c a l l y definab l e , t hat i s it o c c ur s in word­
final po s i t i o n , whi l e [kJ ( t he ot her al l ophone of t he phoneme /k/)
o c c u r s in wor d - i ni t i al and wor d -medial p o s i t ions ( s ee e . g . Panitia
Edjaan Lembaga Bahasa dan Kesusas teraan ( The S pe l ling Comm it t ee o f t he
In s t i t ut e of Language and L i t er a t ur e ) 1 96 6 :5 - 7 ; Johns 1 97 5:8 - 1 7, et c . )
A s far a s c l a s s i c a l phoneme t heory i s c on c erne d , one has t o t a ke
t he f i r s t po s i t i on b ec au s e t he two c on s o nant s ound s ( [kJ and [?J ) do
c ont r a st i n a few pai r s o f l e xic al i t em s , s u c h as:

( 2. 1 ) pak [pakJ ' paa kage ' vs pak [pa?J ' fa t he r , sir '
bak [bakJ ' tr o ug h ' vs bak [ba?J ' l i k e , as '
sukun [sukonJ ' brea dfrui t ' v s suun [su?onJ ' ri a e n o o dl e '
sakat [sakatJ 'para s i t i a p l ant ' vs saat [sa?atJ ' mo m e n t , t ime '

N o t e however t hat t he f i r s t t wo it ems l i s t e d on t he l e f t in t he abo ve

exampl e s are l o anwor d s . In genera l , one c an say t ha t t he ve lar s t o p
([kJ ) o c c ur s in word -fina l po s it i on only in loanwo r d s . I t w i l l b e s e en
l at er t hat loan it em s , part i c ularly t ho s e t hat cont a in for e i gn s o und s
(i.e. sounds wh ic h do not o c c u r in nat ive B I i t em s ) , have a l so b ee n t he
main sour c e o f d i ff i cu l t i e s in s e t t ing fort h und er l y i ng con sonant s and
und erl y ing repre sent at i o n o f morpheme s . The d i ffic u l t i e s ar i s e b e c au s e
t he fore ign s ound s t en d t o b e s ub s t i t ut e d by some ot he r sounds whic h
o c cur in nat ive B I it em s , and ac c or ding l y , most l o an i t em s t en d t o have
t wo or more phonet i c a lt ernant s e ac h .
One fundament al c o nc ept i on i n gener at ive phonol ogy i s t hat e a c h
morpheme ha s o n e and only one b as e form in t h e under lying r e pre s e nt at io n ,
idea l l y one of t he o c c urring phone t i c forms , from whi ch a l l ot her vari­
l
ant s t hat o c c u r in d i fferent environment s are t o b e derive d b y rule s .
The c ho i c e o f a c ert a i n phone t i c form r at he r t han t h e o t h e r s a s t h e
under l yi ng r e pr e s e n t at i o n o f a morpheme m u s t ne c e s s a r i l y b e made o n a
pr inc ipl e d b a s i s t o t he e x t ent t hat suc h c h o i c e has a s impl i fy ing e ff e c t
o n t he grammar o f t he language i n que st i on .

l
c f . Chomsky and Halle 1968 : 9-12 , 164-166 ; Ander son 1974 : 43 ,47 ; Schane 1973 : 74-8 3 ;
Hyman 1975 : 8 0-82 , etc .
14

H a l im ( 1 9 7 4 ) in his b r i e f generat ive treatment o f t he s e gme n t a l

l
phono l ogy of B I po stulat e d a l l t he 2 3 c ons onant s as underly ing s e gment s .
The nature and purpo s e o f h i s de sc r i pt ion ( wh i c h was o r i ginal ly i nt en­
ded as an append i x o f h i s Ph . D . t he s i s ent it l e d ' Int onat i o n in Re l at i o n
t o S y ntax in B aha sa I nd one s i a ' ) al low h i m t o b e f r e e from t he t a s k o f
pr ovid ing some j us t i fi c at ion f o r t he po st u lat ion of t ho s e under lying
c o n so nant segment s apart from a phonemic invent ory . Nor doe s he suggest
any pro c edure or c r i t er io n for t he choic e o f a part i cu la r phone t i c al­
t ernant a s t he underlying form o f a morpheme in t he c a se o f l oan i t ems
t hat have t wo or more phonet i c variant s eac h .
Let u s now return t o t he main que s t i o n . What c on sonant s e gment s do
we need t o po s t u l a t e a s unde r l y ing con sonant s for t he phono l o gi c a l
de s c r i pt ion o f B I , so t hat t he d e s cr ipt ion c apture s max imum general­
i sat i o n s o f t he language? To put it in a d i fferent way , do a l l t he 2 3
c on sonant segment s have t o b e po s t u lat e d a s und e r ly ing s e gment s i n order
to ac h ieve t he b e st r e sult ? If t he s ame r e su l t c an b e obtained w i t h
l e s s und e r l y ing segment s , it i s sure ly d e s irab l e t o po st ulat e l e s s s e g ­
ment s for t he under l y ing repre s ent at io n .
As ha s been ment ioned ear l ie r in t hi s se c t io n , t he pro b l ems fa c e d
i n s e t t ing up und e r l y ing c on sonant segment s revolve around l o an i t e m s
a n d f o r e i g n sounds whic h t h e l oan i t e m s have brought a l o n g int o t he
sound s y s t em o f t he language . For t h i s part i c ular rea son , t he di s c u s s ion
in t he paragraphs t o fol low wi l l make a d i s t inct ion b e t ween indigenous
sound s ( i . e . sound s whi c h o c c ur i n nat ive BI wo rds ) and l o an sound s .
Spec i f i c a l l y , t he fri c at ive s [f , 5, z, xJ w i l l b e d i s c u s s e d separat e ly
from t he rema i n ing 1 9 consonant s . The four fric at ive s ( [ f , 5 , z, xJ ),
wh i c h are loan sound s , w i l l be d i s c u s s e d a f t er t he 19 c o nsonant s .

2 . 1. 1. T H E C O NS O NANT S YS T E M I N I NV I G E NOUS W O R VS

A s ha s a lr eady been indic at e d in Chart 1 above , t he 1 9 consonant

sounds whi c h o c cur in nat ive B I ( Malay ) wor d s cons i s t of s e ven plo s ive s
( st ops ) ( [p , b, t , d, k , g, ? J ) , two fri c at ives ( [ 5 , h J ) , two a fr i c a t e s
([�, J J ) , four n a s a l s ( [ m , n, fl, I) J ) , t wo l iqui d s - o n e l a t e r a l and one
t r i l l - ( [1 , r J ) , and two semivowe l s or g l i d e s ( [w, yJ ) .
Funct i o na l l y , t he 1 9 c o n s onant sounds can b e c la s s i fied into 1 8 d i s ­
t in c t ive sound unit s known a s phoneme s . The s e 1 8 c o n sonant phoneme s

l
This is the only generative treatment of BI phonology to my knowledge . The descrip­
t i on however is not detailed enough for it is limited to the level of format ive and
it account s for six processes only, i . e . vowel nasalisat ion, schwa deletion , vowel
laxing , voic ing of /h , ? /, deletion of /h/ and nasal-obstruent as similat ion.
 15

are a l l p o s t u l a t e d a s und e r l y ing c o n s on ant s e gme nt s and wri t t en b et ween

slashes (I I) in informal repre sent at ion o f morpheme s . They are Ip, b,
t, d, k, g, s, h, c, J, m, n, p, I), I, r, wi and Iy /. N o t e t hat symb o l s
and n ame s u s e d for t he und e r l y ing ( p hono l o g i c a l ) con sonant s e gment s
ar e s imi l ar t o t ho se of the p honet i c repre s entat i o n .
B e fore we e x amine t he re l at i o n s hi p s between t h e unde r l y ing s e gment s
w i t h regard t o e a c h ot her ( i . e . t he i r d i st inct i v e re l a t i o n ) and b e t w e e n
t he unde r l y ing c o n s onant s a n d t h e i r r e spe c t ive phone t i c r e a l i s at io n , i t
i s nec e s s ary t o make a short remark on t he nature o f t he de s cript i o n .
The analys i s made i s mainly on t h e l e v e l o f morpheme . Pro c e s s e s t hat
i nv o l ve und e r l y ing s egment s acro s s morpheme b oundar i e s , su c h a s con­
sonant d e l et i o n , et c . are however t aken int o a c c ou nt . In order t o
e s t ab l i sh t he i de nt i t y o f a p honeme ( i . e . it s int ri n s ic i dent i t y whic h
shows t hat it i s d i fferent from the o t her p honem e s ) a s ep arat e minimal
pair for every opp o s e d phoneme i s given .

2 . 1 . 1. 1 . The Underl ying Consonant /p/

T h i s underly ing c o n s o nant may o c cur in morp heme i n i t ial , me d i al and

final p o s i t i o n s . I t s ident it y and di st i n c t i v e func t ion c an b e e st a b ­
l i shed b y t he fo l lowing compar i so n :

( 2 . 2) a. p/b: /paku/ [paku] ' na i l ' - /baku/ [baku] ' r aw ma t er i a l '

b. pit : /panah/ [panah] 'arr ow ' - /t anah/ [t anah] ' l and, s oi l '
c. p/d: /apa/ [apa] ' w ha t ' - /ada/ [ada] ' exi s t '
d. p/k: /paral)/ [paral)] ' knife ' - /karal)/ [karal)] ' r oc k s '
e. p/g: - /garal)/ [garal)] ' grim '
f. pis: - /saral)/ [saral)] 'nes t '
g. p/h: /lapar/ [lapar] ' hungry ' - /lahar/ [lahar] ' la v a '
h. pic: /kupil)/ [kUPLI)]1 'ear ' - /kucil)/ [kUCLI)] ' oa t '
1. plj: /padi/ [padi] 'rice ' - Ijadi/ [jadi] ' b e c ome '
j. p/m: /palu/ [palu] ' hamm e r ' - /malu/ [malu] ' s hy '
k. pin: /padi/ [padi] 'rice ' - /nadi/ [nadi] 'pulse '
1. p/p: /paman/ [paman] ' u nc l e ' - /paman/ [paman] ' hea l t hy '
m. p/I): /sadap/ [sadap] ' de l i c i ou s ' - /sada)
l / [sadal)] 'me dium '
n. p/l: /puas/ [puas] ' sa t i s fi e d ' - /luas/ [luas] ' w i de '
o. p/r: /puas/ [puas] ' sa ti s fi e d ' - /ruas/ [ruasJ ' j oi n t '
p. p/w: /lipat / [lipat ] ' f ol d ' - /liwat / [liwat ] ' t o pass '
q. ply : /sapa/ [sapa] 'gree t ' - /say a/ [say a] 'I'

The underly ing c o n so nant /p/ i s p honet i c a l l y real i s e d as [p] and

produc ed w it h l o wer lip and upper lip a s a c t ive and p a s s ive art i cu lat o r s
r e sp e c t i v e l y and w i t hout t he v ibrat i o n o f t h e v o c a l c ords . In final
po s i t i o n [ p ] i s pronounc ed w i t hout any plo s i on . But when / p / o c curs
a ft er t he pre fix /ma n / ' ac t ive voic e verb pre fix' ( i . e . w hen /p/ is a
s t em- in i t i a l e l ement ) t hen it w i l l be phonet i c al ly real i s e d a s [ �] ,
suc h as: I m a n
' to u s e ,
+ p u ku l l [ m a m u ko l ]
to wear ' , e t c .
' to bea t ' , I m a n
( see S e c t io n 5 . 5 .
+ p a kayl [mama k a y ]
for a furt her d i sc u s s io n o f
t hi s phenomeno n ) .

2. 1 . 1.2 . The Underlying Consonant Ibl

T h i s und e r l y ing c on sonant may o c c ur in morpheme init i a l and me d i a l

po s i t i o n s o n l y . I t s id ent i t y and d i s t inct ive fun ct ions c an b e e st a b ­
l i shed b y t h e fo l l ow i ng c ompari son:

(2.3) a. b/p : ( s ee 2 . 2:a )

b. b i t : li b u l [ ib u ] 'mo the r ' - li t u l [ i t u ] ' tha t '
c. b i d : la b u l [ a b u ] ' a s h ' - la d u l [ a d u ] ' to c l a s h '
d. b/ k : - la k u l [ a k u ] 'I '
e. big : I b i l a l [ b i l a ] ' w he n ' - I g i l a l [ g i l a ] ' araz y '
f. b/ s : - / s i l al [ s i l a ] ' pr i n a i p l e '
g. b / h : I b i l a !)1 [ b il a !) ] ' te H ' - I h i l a !)1 [ h i l a !)] ' lo s t '
h. b / e : I b a r a !)1 [ b a r a !)] ' g oods ' - I ca r a !)1 [ e a r a !)] ' twi g '
1. b/J : - / Ja r a !)1 [ l a r a !)] ' rare '
j. b/m : I b ua t l [ b u a t ] ' do, ma ke ' - I m u a t l [ m u a t ] ' to l oa d '
k. b i n : I b a i k l [ b a L ? ] ' g o o d ' - I na L k l [ n a i ? ] ' g o up '
1. b/p : Iba tal [ b a t a ] ' br ia k ' - I p a t a l [ p a t a ] ' re a l , a l ear '
m. b / !) : I b er i l [ b a r i ] ' g i v e ' - I !)e r i l [ !)a r i ] ' to s h udder '
n. b/ l : I b u a s l [ bua s ] ' wi l d ' - I l ua s l [ l u a s ] ' wi de '
o. b lr : - I ru a s l [ r ua s ] ' jo i n t '
p. b/w : I l i ba t l [ l i ba t ] ' i nv o l v e ' - / l iwa t l [ l i wa t ] ' to pa s s '
q. b/y : I I a bul [ I abu] ' p um p k i n ' - / I ayul [ I ayu] ' w i the r e d '

The und e r l ying c on sonant I b l i s phone t i c a l l y real i s e d a s [ b ] prod u c e d

wit h l o wer l i p and upper l ip a s a c t ive and pa s s ive art i c u la t ors re spe c ­
t iv e l y and t h e v o c a l c ords are i n vibrat ion .

2.1. 1 . 3. The Underlying Consonant It I

T h i s underly i ng c o n s onant may o c c ur in morpheme i n i t i a l , me d i a l and

final po s i t i on s . It s ident i t y and d i st inc t ive fun c t ion c an b e e st ab ­
l i shed b y t he f o l lowing c ompar i s ons .

( 2 . 4 ) a. t i p : see ( 2 . 2 : b )
b. t / b : s e e ( 2 . 3:b )
c. tid : I t u a l [ t ua ] ' o ld ' -
Idual [ du a ] ' two '
d. t / k : I s a tul [ sa t u] 'one ' - I s a k u l [ s a k u ] 'po a k e t'
 17

e. t/g : - I s ag u l [ s a g u ] ' p a lm - l i k e t r e e '

f. t i s : la t a p i [ a t a p J ' ro o f ' - la s a p l [ a s a p J ' sm o k e '
g. t/h : I ta r i l [ t a r i J ' da nc e ' - I h a r i l [ ha r i J ' day '
h. tic : - lea r i I [ e a r i J ' to l o o k for '
i. t/J : /Ja r i I [Ja r i J ' finger '
j. tim : - I m a r i l [ ma r i J ' co m e on '
k. t i n : I m a t a l [ ma t a ] 'eye ' - Imanal [ma n a ] ' w h e r e , w hi c h '
1. tip : I t a ma n l [ t ama n J ' p a r k ' · - I p am a n l [ p a m a n ] ' he a l t h y '
m. t i l) : I b u t a l [ b u t a J ' b l ind ' - I b u l) a l [ b u l) a ] ' flower '
n. t / 1 : I t u a l)l [ t u a l) ] ' p o ur ' - I l u a l) l [ l u a l) J ' spare ( tim e ) '
o. t l r : I t u a l) l [ t u a l) ] ' p our ' - I r u a l) l [r u a l) J ' ro om '
p. t /w : la t a s l [ a t a s J ' top, on ' - law a s l [ a w a s J ' b e c a r e fu l '
q. t / y : I ba t u l [ ba t u ] ' stone ' - I bayul [ bayu] ' wind '

T h e underlying c o n sonant I t I i s bas i c a l l y rea l i se d as [ t J o n t he

p honet i c leve l . Thi s c on so nant i s a vo i c e l e s s una s p irat e d ap ic o -d e n t a l
( or ap i c o -d ent i a lveolar ) sound . In final p o s i t ion , it i s pronoun c e d
wit hout any p l o s i on . Some speaker s , under t he in fluence o f t he ir
regional language, u s e a r e t r o f l e x ed [ t J but t h i s var iant i s con s idered
non-st andard .
The under l ying consonant I t I can be rea l i s e d as [ 0 J on t he phone t i c
l eve l , i . e . when i t i s a s t em - in i t i a l c on sonant and o c cur s after t he
pr e f ix /m a n / , s u c h as: I m a n + ta r i l [ m a n a r i J ' to dan ce ' . Iman + t u l i sl
[manu l i s J ' to wr i t e ' , e t c . ( See a furt her d i s c u s s i on of t h i s p henomenon
i n S e ct ion 5 . 5 . )

2. 1. 1 . 4. The Underlying Consonant / d /

Thi s under lying c on sonant may o c c ur i n morpheme init i a l and med i a l

p o s it io n s onl y . I t s ident i t y and d i st in c t ive func t ion can be e st ab­
l i shed by t he fo l l ow ing compar i so n s :

( 2 . 5) a. d i p : see ( 2 . 2 : c )
b. d / b : see ( 2 . 3 : c )
c. d / t : see ( 2 . 4 : c )
d. d / k : Idudal [dudaJ ' w i dower ' - I d u kal [ du k a J ' s ad '
e. dig : -
Idugal [ duga] 'guess '
f. dis: Idukal [ dukaJ 'sad' - I s u kal [ su ka] ' to l i k e '
g. d / h : I da r i I [ d a r i J ' from ' - I h a r i l [ ha r i J ' day '
h. die : -
le a r i l [ e a r i ] ' to l o o k for '
i. d/J : - /Jar i l [Ja r i J ' fing e r '
j. dim : Ida r i l [ d a r i J ' from ' - I m a ri l [ ma r i J 'c ome on '
k. d i n : I d ad a l [ d a d a J 'che s t ' - I d a n a l [ d a na J ' funds '
18

-
l. d I p : I s ud i l [ s u d i ] ' wi t t ing ' I s u p i l [ s up i ] ' qu i e t '
m. d / !) : I s u d u t l [ s u d o t ] ' corner ' - I s u l}u t l [ S U !) o t ] ' g r um b t e '
n. d/ 1 : Idar i I [ dar i ] , from ' - I Iar i I [ I ar i ] ' ru n '
o. d l r : I d a "d a l [ d a d a ] ' che s t ' - I da r al [ d a r a ] ' y oung woman '
p. d / w : / l a d a !) 1 [ I a d a !) ] ' fi e t d ' - I l a w a !) 1 [ I a w a !) ] 'gate, do o r '
q. d/y : Idadal [ dada] 'chest ' - I da y a l [ da y a ] ' power '

The u nderly i ng c o n s o nant I d l i s phone t i c al ly real i s e d as C d ] . This

c on so nant i s a voic e d api c o -d ent a l ( or apic o - dent i alveolar ) s ound .
Again , t hi s sound i s often r e t r o f l e x e d b y some speake r s . S uc h r e a l i sa­
t ion however i s c on s id ered non-st andard .

2 . 1. 1 . 5 . The Underl ying Consonant /k/

T h i s und e r l y i ng c o n sonant may o c c ur in morpheme init ial , medial and

final po s i t i on s . I t s ident i t y and d i st i n c t ive funct i on c an be e st ab ­
l i s he d b y t he fo l lo wing c ompari son s :

(2.6) a. k I p : see ( 2 . 2 : d )
b. k/b : see ( 2 . 3 : d )
c. k I t : see ( 2 . 4 : d )
d. k I d : see ( 2 . 5 : d )
k / g : la k a r l [ a k a r ] -
e. ' ro o t ' I ag a r I [ a g a r ] ' in o r der '
f. k/ s : I k u k u l [ ku ku ] ' nai t ' - / s u ku l [ s u k u ] ' e t hn i c group '
g. k / h : la r a k l [ a r a ? ] ' to pro ce s s ' - la r a h I [ a r a h ] ' di re c t i on '
k / c: I k u r a !) 1 [ k u r a !) ] -
h. ' te s s ' le u r a !) 1 [ cu r a !)] ' to cheat '
k/J : -
1. / ju r a !) 1 [ J u r a !) ] 'gorge '
k I m : I k a t a l [ ka t a ] -
j . ' wor d ' I m a t a l [ ma t a ] 'eyes '
k. kIn : / pa k a k l [ p a k a ? ] ' de a f ' - I paka n l [ p aka n J 'week '
l. k I p : I ka t a l [ k a t a J -
' wo r d ' Ipatal [pataJ 'rea t , c te a r '
m. k / !) : la k a n l [ a k a n ] ' wi t t ' - la !) a n / [ a lla n ] ' t ho ug h t '
n. k lI : I l a ku l [ I a k u ] ' s o td ' - I I a I ul [I a I uJ 'pa s t '
-
o. k lr : I ka y a l [ k a y a J ' ri ch ' I rayal [ rayaJ ' t arge '
p. k / w : I ka t a k l [ ka t a ? J ' frog ' - Iwa t a k l [wa ta ? J ' c harac t e r '
-
q. k lr: I b a k a r l [ b a ka r J ' b urn ' I b a y a r l [ b ay a r J ' to pay '

The und e r l y ing c on s o nant I k l i s b a s i c a l l y r e a l i s e d as a voi c e l e s s

ve l ar s t o p ( [k J ) i n morpheme init ial and me dial po s i t i on s , and as a
( vo i c e l e s s ) g l o t t a l s t op ( [ ? J ) in morpheme fina l po s i t ion . ( See a
fur t he r d i s c u s s ion o f t he di stribu t i o n of t he se s ound s in S e c t ion 5 . 2 . )
When I k l i s a st em-in i t i a l segment and it o c c ur s a ft er t he pre fi x /m a n / ,
t hen i t w i ll b e phonet i ca l l y r e a l i s e d a s [ 0 J , s u c h as: I m a n + k a t a +

k a n l [ m a !) a t a k a n J ' t o t e t t, say ' , Ima n + ka s i h l [ m a !) a s i h J ' to give ' , etc .

( S e e S ec t ion 5 . 5 . for a furt her d i s c u s s ion o f t hi s phenomenon . )
 19

2.1.1.6. The Underlying Consonant Igl

T h i s underly ing c o nso nant may o cc ur in morpheme init i a l and me d i a l

p o s it ions only . It s ident i t y and d i s t i n c t i ve func t ion c an b e e s t ab ­
l i shed by t he fo l l owing c ompa r i s on s :

(2. 7) a. g/p : see ( 2 . 2 : e )

b. g/b: see ( 2 . 3 : e )
c. 9It : see ( 2 . 4 : e )
d. g / d : see ( 2 . 5 : e )
e. g / k : see ( 2 . 6 : e )
f. g/s : I ragal [ ragaJ ' b ody ' -
I ra s a l [ ra s a J ' fe e l i n g '
g. g/ h : I b a g a n l [ b a g a n J ' de s i g n ' -
Ibahanl [bahan] 'ma t er ia l '
h. g / c : la g a r l [ a g a r J ' in order ' -
la ca r l [ a ca r J 'pickle '
i. g/J : la g a r l [ a g a r J ' in order ' -
la J a r l [ a J a r J ' t ea c h '
j. g /m : I n a g a l [ na g a J ' dr ag on ' -
I n ama l [ n am a J ' name '
k. gin : Ig i a tl [ g i a t J ' di l i g e n t ' -
In i a tl [ n i atJ ' i n te n t i on '
l. g/p : Igar i sl [ g a r � s J ' line ' -
I p a r i s l [ p a ns J ' ne ar l y '
m. g i l) : I t a g a h l [ t a g a h ] ' forb i d ' -
I t a l) a h l [ t a l) a h J ' ha lf '
n. g / 1 : I l agul [ I agu] ' s ong ' -
l1 a l u l [ I a I u J ' pa s t '
o . g lr : 1 1 a g i I [ l ag i ] ' again ' -
/ 1a r i l [ I a r i J ' run '
p. g /w : I r a g a l [ r a g a J ' b ody ' -
I r aw a l [ r a w a J ' swamp '
q. g/y: -
I rayal [ ra y a J ' la r g e '

The underlying c o n sonant I g l i s p h on e t i c a l ly r e a l i s e d as a voic e d

v e l a r s t op ( [ g J ) .

2.1.1.7. The Underlying Consonant l si

This underly ing c o n so nant may oc cur in morpheme i n i t i a l , medial and

final p o s i t ion . I t s ident it y and d i s t inct ive func t ion c an be e s t ab ­
l i shed b y t h e fo l l o w ing c ompar i sons :

( 2. 8 ) a. s i p : see ( 2 . 2 : f ) b. s i b : see ( 2 . 3 : f )
c. s lt : see ( 2 . 4 : f ) d. s i d : see ( 2 . 5 : f )
e. s lk : s e e ( 2 . 6 : f ) f. s I g : s ee ( 2 . 7 : f )
g. s/h : I sa r i l [ s a r i J ' e s s enc e ' -
I h a r i I [ ha r i J ' day '
h. s lc: -
I ca r i l [ ca r i J ' t o l ook f or '
i. s/J : -
/ J a r i l [ ja r i J ' fi n ge r '
j . s/m : -
Ima r i I [ ma r i J ' c om e on '
k. s i n : I ka s a l l [ kasa l J ' ann oy e d ' -
I kan a l l [ ka n a l J ' fam i l iar '
l. sip : -
I k a p a l l [ k ap a l J ' e las tic '
m. s / l) : I l u a s l [ l u a s J ' w i de ' -
I l u a l) l [ I u a l) ] ' spare ( t ime ) '
n. s / 1 : law a s l [ aw a s J ' ca r e fu l ' -
lawa l l [ aw a l J 'b e g i nning '
20

o. si r : I l u a s l [ l ua s J ' w i de ' -
/ l ua r l [ l ua r J ' ou ts i de '
p. s /w : I d e s a l [ d e s a J ' v i l lage / -
I d e w a l [ d e wa J ' g od '
q. s ly : I s a r) 1 [ s a r) J ' th e ' -
ly a r) 1 [ y a !) J ' w hi c h '

The under l y i ng c on s on ant ls i i s b a s i c al ly r e al i se d as a vo i c e l e s s

lamino a lveolar fr i c at ive sound ( [ s J ) . But when t he underl y ing ls i i s
a s t em-init i a l c on sonant and o c cu r s after t he pr e fix / m a n / t he n i t w i l l
b e p ho ne t i c a l l y re a l i s e d a s [ 0 J , s u c h a s : I m a n + s a p u l [ m ap a p u J ' to
s w e ep ' , I ma n + s i s i r l [ ma p i s i r J ' to c omb ' , e t c . ( Se e a furt he r d i s ­
c u s s i on o f t hi s phe nomenon i n S e c t i on 5 . 5 . )

2. 1. 1. 8 . The Underlying ( Consonant ) G l ide Ihl

T h i s und e r l y ing gl i d e may o c c ur i n morpheme i n i t i a l , m e d i al and

f i na l p o s it i o n s . I t s ident i t y and d i st inct ive func t i on c an b e e st a b ­
l i sh e d by t he foll owing c ompar i sons :

( 2 . 9) a. h lp : s e e ( 2 . 2 : g ) b. h / b : see ( 2 . 3 : g )
c. h i t : see ( 2 . 4 : g ) d. h i d : see ( 2 . 5 : g)
e. h / k : see ( 2 . 6 : g ) f. h / g : see ( 2 . 7 : g )
g. hi5: see ( 2 . 8 : g )
h. hie : I ha r i I [ ha r i J ' day / -
I �a r i I [ � a r i J ' to l ook for /
i. h!j: -
!jar i l [ Ja r i J ' fi n ge r '
j . him : -
Im a r i l [ ma r i J / c ome on /
k. h/n : I tuahl [ tuahJ / luck / -
I t ua n l [ t u a n J /mas ter /
l. h lp : I h a w a l [ h a w a J 'air / -
I p a w a l [ p awa J / s ou l /
m. h i !) : I b u a h l [ b u a h J / fr ui t / -
I b u a r) 1 [ b u a r) J ' to th r ow /
n. hi I: -
I b ua l l [ bu a l J ' b oa s t /
o. h ir : Idah i l [ dah i J ' f or e he a d / -
Ida r i I [da r i J ' fr om /
p. h/w : I t a h a n l [ ta h a T J ' de ta i n / -
I t a wa n l [ t a w a n J / cap ture /
q. h / y : I l a h a r l [ l a ha r J ' lava / -
/ l a y a r l [ l a ya r J / sa i l /

The und e r l y ing g l ide I h l i s b a s i c a l l y real i sed as a voic e l e s s g l o t t a l

fri c at ive sound ( [ h J ) . When I h l oc c ur s i n intervo c a l i c p o s it i o n , i t
t end s t o b e phone t i c a l ly real i s e d a s a v o i c e d glot t a l fr i c at ive sound
( [ hJ ) .
The under l y ing I h l c an be p honet i c a l l y rea l i s e d a s [ 0 J when i t
oc curs i n morpheme final p o s i t i on , s u c h a s I r um a h l [ r u m a h , ruma J ' h ou s e ' ,
I s ud a h l [ s u d a h , s ud a J ' fin i s he d ' , e t c . , or when it o c c ur s b etween two
non - id ent i c a l vowe l s i n ver b s and adj e c t iv e s , su c h a s I I i h a t l [ I i h a t ,
l i a tJ ' to s e e ' , I p a h i t l [ p a h L t , p a L tJ ' b i tter / , e t c . ( S ee a furt her
d i s c u s s i on o f t he z er o rea l i sa t i o n of I h l in S e c t ion 5 . 3 . )
 21

2. 1 . 1 . 9 . The Underlying C onsonant /�/

This und erly ing c on s onant may o c cur in morpheme i n i t i a l and me d i a l

p o s it i o n s only . I t s ident i t y and d i s t in c t ive fun c t i on c an b e e st ab ­
l i s hed b y t he fo l l owing compar i so n s :

( 2 . 10 ) a . C lp : s e e ( 2 . 2 : h ) b. C lb : s e e ( 2 . 3 : h )
c. C i t : see ( 2 . 4 : h ) d. Ud : see ( 2 . 5 : h )
e. U k : see ( 2 . 6 : h ) f. Ug : s e e ( 2 . 7 : h )
g. C i s : see ( 2 . 8 : h ) h . C lh : s e e ( 2 . 9 : h )
i. C lj : I a c a r I [ a e a r ] ' p ic k l e ' - la j a r l [ a j a r ] , t e ac h '
j. e lm : I c i n t a l [ c i n t a ] , l ov e ' - I m i n t a l [ m i n t a ] 'ask ( f or ) ,
k . U n : I k u c i r) 1 [ k U C L r) ] ' e a t ' - I k u n i r) 1 [ k u n L Q ] ' y e l l ow '
l . U p : I s u c i l [ s u c i ] ' h ol y ' - I s u p i l [ s u p i ] ' q uie t '
m. U r) : I p a c a t l [ p a e a t ] ' di s m i s s ' I p a r) a t l [ p a r) a t ] ' s t e w '
n. C ll : leo b a l [ c o b a ] ' t ry ' - I l o b a l [ l o b a ] ' g r e e dy '
o. C lr : I c u c i l [ c u e i ] ' wa s h ' - I c u r i l [ e u r i ] ' s te a l '
p. c/w: I bacal [ ba c a ] ' re a d ' - I b aw a l [ b a w a ] ' b r i ng '
q. U y : I ka c a I [ ka c a ] ' g la s s ' - I k a y a l [ k a v a ] ' ri c h '

The unde r l y i ng c on sonant I c l i s p ho net i c a l l y r e a l i s e d a s a voic e l e s s

lamino alveo l o -p a lat a l affr i c at e ( [ c ] ( = I PA [ t Q ] ) ) .

2 . 1 . 1. 10 . The Underl ying Consonant / 3 /

T h i s under ly ing c on sonant may o c c ur i n morpheme i n i t i a l a n d m e d i a l

p o s i t io n s only . I t s ident i t y and d i st inc t ive fun c t i o n c an be e st a b ­
l i shed b y t he fo l lo wing c ompa r i s on s :

( 2 . 11 ) a . J / p : see ( 2 . 2 : i ) b. J i b : see ( 2 . 3 : i )
c. J I t : see ( 2 . 4 : i ) d. l i d : see ( 2 . 5 : 1 )
e. J / k : see ( 2 . 6 : i ) f. J ig : see ( 2 . 7 : i )
g. J / s : see ( 2 . 8 : i ) h. J / h : see ( 2 . 9 : i )
i. J / e : see ( 2 . 1 0 : i )
j . J i m : lJ a r i l [ J a r i ] ' fi n g e r ' - I ma r i l [ m a r i ] ' c om e on '
k. J / n : lJ a d i l [ J a d i ] ' b e c om e ' - I n a d i l [ n a d i ] ' p u l s e '
l. J / p : I p u J a l [ p u J a ] ' a d or e ' - I p u p a l [ p u J1 a ] ' p os s e s s '
m. J I r) : I h a J a t l [ h a J a t ] , de s ire ' - I h a r) a t l [ h a r) a t ] ' w ar m '
n. J I l : / j a r i I [ J a r i ] ' fi n g e r ' - 1 1 a r i I [ 1 a r i J ' r un '
o. JI r : IJasal [ J a sa] ' se r v i ce ' - I r a s a l [ r a s a ] ' fe e l '
p. J /w : I s a j a l [ s a J a ] ' j u s t ' - I s aw a l [ s aw a ] ' py t h on '
q. J/y : - I s ayal [ saya] 'I'

T he under l y ing c o n sonant I J I i s p honet i c a l l y rea l i sed a s a v o i c e d

lamino a l v e o l o -p a l at a l a f fr i c at e ( [ J ] ( = IPA [ d ¥ ] ) ) .
22

2 . 1. 1 . 11 . T h e Unde rly i n g C o n s o n a n t Iml

T h i s under l y i ng c on s onant may o c c u r in morpheme i n it i a l , med i a l and

f i nal p o s i t i on s . I t s i dent i t y and d i st in c t ive func t i on can b e e s t a b ­
l i shed by t h e fo l l owing c ompar i so n s :

( 2 . 12 ) a. m / p : see ( 2 . 2 : j ) b. m / b : see ( 2 . 3 : j )
c. m lt : s e e ( 2 . 4 : j ) d. mi d : see ( 2 . 5 : j )
e. m / k : see ( 2 . 6 : j ) f. mi g : see ( 2 . 7 : j )
g. m / s : see ( 2 . 8 : j ) h. m/ h : see ( 2 . 9 : j )
i. m / c : see ( 2 . 1 0 : j ) j . m / J : see ( 2 . 1 1 : j )
k. m i n : I s am a l [ s a m a ] ' s ame ' - I s a n a l [ s a n a ] ' th e r e '
l. m / p : I h amal C h ama ] ' di s e as e ' - I hapal C h apa ] 'only '
m. m i l) : I g a r a m l [ g a r a m ] , sa l t ' - I g a r a l) l [ g a r a l) ] ' fi e r c e '
n. mlI : I m a l u l [ ma l u ] ' shy ' - I I a l u i L l a l u ] 'pas t '
o. m / r : Ima s a l [ ma s a ] ' p e r i o d ' - I r a s a l [ r a s a ] ' fe e l '
p. m / w : la m a n l [ a m a n ] ' s afe ' - law a n l [ a w a n ] ' c lo u d '
q. m/y : I samal [ sama ] ' s am e ' - I s a y a l [ s a y a ] 'I'

T h e und er ly i ng c o n so n ant I m l i s p honet i c a l ly r e a l i s e d a s a vo i c e d

b i l ab i a l na s a l s ound ( [ m ] ) .

2 . 1 . 1 . 12 . The Und e rl y i n g C o n s o n a n t In l

T h i s under l y ing c on s o nant may o c c ur in morpheme i n it i a l , me d i a l and

f in a l p o s i t i o n s . It s ident it y and d i s t i n c t i ve func t io n c an be e st ab ­
l i s h ed b y t he fo l l ow i ng c ompar i sons :

(2. 13) a. n i p : see ( 2 . 2 : k ) b. n i b : see ( 2 . 3 : k )

c. n i t : see ( 2 . 4 : k) d. n / d : see ( 2 . 5 : k)
e. n lk : s e e ( 2 . 6 : k ) f. n / g : see ( 2 . 7 : k )
g. n / s : see ( 2 . 8 : k) h. n i h : see ( 2 . 9 : k )
i. n / c : see ( 2 . 1 0 : k) j . n ! j: see ( 2 . 11 : k )
k. n / m : see ( 2 . 12 : k)
l. n i p : I k a n a l) l [ k a n a l) ] ' re c a l l ' - I k a p a l) l [ k ap a l) ] ' fu l l ( s tomac h ) ,
m. n / f) : la s i n l [ a s L n ] , s a l ty ' - la s i f) 1 [ a S L f) ] ' s trange '
n. n i l : I n am a l [ n a m a ] ' name ' - I l a m a l [ l a m a ] ' long ( time ) ,
o. n / r : la k a n l [ a k a n ] ' w i l l ' - I a ka r I [ a k a r ] 'roo t '
p. n/w: Idenal [dena] ' de s i g n ' - I d ew a l [ d ew a ] 'god '
q. n/y : Isanal [ sana] ' the r e ' - I s a y a l [ s a y a ] 'I'

The und er lying c on s o nant I n l i s b a s i c a l ly real i sed on t he phonet i c

l e v e l a s a vo i c e d ap i c o dent al ( or ap i c o dent i - a l v e o l a r ) na s a l sound
( [n] ) . The u nd e r l y ing I n l i n t he p r e f i x I m a n l is p hon e t i c a l ly rea l i s e d
a s a v o i c e d b i l ab ial n a s a l ( [ m ] ) b e fore b il a b i a l p l o s i ve s suc h a s 1 m a n +
 23

b a c a l [ m am b a c a J ' t o read ' , I m a n + p i l ih l [ m amil ih J ' t o c h oose ' , e t c . ;

as a vo i c ed l am ino a l v e o l o -p a l a t a l nasal ( [ p J ) b e fore lamina l fr i c at ive s
and a f fr i c a t e s suc h a s I m a n + s a p u l [ m a p a p u J ' t o sweep ' , I m a n + ca r il
[ m a p ca r iJ ' t o Z ook f or ' , et c . ; a s a vo i c e d dor s o v e l ar na s a l ( [ 8 J )
b e fore velar and g l o t t a l o b st r uent s and b e for e vowe l s suc h a s I m a n +

g a 8g ul [ma 8g a 8 g u J ' t o d i s t ur b ' , I m a n + h u k u m l [ m a 8 h u kom J ' t o p un i s h ' ,

Iman + a J a r l [ma 8a J a r J ' t o t e a c h ' , e t c . ; and a s [ �J b e fore sonorant
c on sonant s such as I m a n + m in t a l [ m am in t a J ' t o a s k f or ' , I m a n + 1 ih a t l
[ m a l ia t J ' t o see ' , Iman + w a k i l + il [ m aw a k il iJ ' t o r e pr e s e n t ' , et c .
( Se e a furt her d i sc u s s i o n of t h i s p henomenon in S e c t i o n 5 . 5 . )

2 . 1. 1 . 13 . The U n derlying Con s on an t / p /

T h i s under l ying c on sonant may o c c ur in morpheme i n i t i a l and medial

p o s i t ions only . It s id ent i t y and d i st inct ive func t io n can b e e st a b ­
l i shed b y t he f o l l owing c ompar i sons :

( 2 . 14 ) a . p I p : see ( 2 . 2 : 1 ) b. p / b : see ( 2 . 3 : 1 )
c. p I t : see ( 2 . 4 : 1 ) d . p/d : see ( 2 . 5 : 1 )
e. p / k : see ( 2 . 6 : 1 ) f. p I g : see ( 2 . 7 : 1 )
I
g. p I s : see ( 2 . 8 : 1 ) h. p / h : see ( 2 . 9 : 1 )
i. p I c: see ( 2 . 1 0 : 1 ) j . p lJ : s e e ( 2 . 1 1 : 1 )
k. p /m : s e e ( 2 . 1 2 : 1 ) l. p I n : see ( 2 . 1 3 : 1 )
m. p / 8 : I p a r il [ p a r iJ 'pain ' 1 8 a r il [ 8 a r iJ ' s hudder '
-

n. p / 1 : I p iu r l [ p ie r J ' c oc onu t ' / l iu r l [ 1 ie r J ' s a Z i v a '

-

o . p Ir : I p a w a l [ p awa J ' s ou l. ' -

I rawal [ ra w a J ' swamp '
p. p /w : I h a p a l C h a p a ] ' on Zy ' -
I h a w a l [ ha w a J 'air '
q. p ly : la r- a m l [ a p am J ' we a v e ' -
la y a m l [ a y a m ] ' f ow Z '

The und e r l ying c on sonant I p l i s p honet i c a l l y rea l i sed a s a vo i c ed

lam ino alve o l o-pa lat al n a s a l sound ( [ p ] ) .

2 . 1.1.14 . The U nderl ying Con sonant / 8 /

T h i s und er l y ing c o n s'onant may o c c ur in morp heme i n i t i a l , me d i a l and

f i n a l p o s i t ion s . It s i d e nt it y and d i st in c t ive func t ion c an b e e s t ab -
l i s hed b y t he f o l lowing c ompar i so n s :

(2. 15) a. 8 / p : see ( 2 . 2 : m ) b. 8 / b : see ( 2 . 3 : m )

c. 8 / t : see ( 2 . 4 : m) d. 8 / d : see ( 2 . 5 : m )
e. 8 / k : see ( 2 . 6 : m ) f. 8 /g : see ( 2 . 7 : m )
g. 8 / 5 : see ( 2 . 8 : m ) h. 8 / h : see ( 2 . 9 : m )
i. 8 / c : see ( 2 . 1 0 : m ) j . 8 1 J : s e e ( 2 . 11 : m )
k . 8 1 m : see ( 2 . 12 : m ) i. 8 / n : see ( 2 . 1 3 : m )
m. 8 / p : see ( 2 . 14 : m )
24

n. 1) 1 I : I t o l o l) l [ t o b l) J ' he lp ' - I to l o l l [ to b l J ' s t up i d '

o. 1) 1 r : I h a l) u s l [ h a l)o s J ' b urn t ' - I h a r u s l [ ha ro s J ' m us t '
p. I) / w : I t a l) a n l [ t a l) a n J 'arms ' - I t a w a n l [ t aw a n J ' t o cap ture '
q. I) / y : I h a l) a t l [ h a l) a t J , warm ' - I haya t I [ haya tJ ' l i fe '

The under l y ing c on sonant I I) I i s p honet i c a l l y r e a l i s e d as a v o i c e d

d o r s o ve l ar n a s a l sound ( [ I) J ) .

2. 1 . 1 . 1 5 . The Underl y i n g Con son ant II I

Th i s underl y ing c o n sonant may o c c ur in morp heme i n i t i a l , m e d i a l and

f i n a l p o s it io n s . I t s ident i t y and d i st inc t iv e fun c t ion c an b e e st ab ­
l i s hed b y t he fo l l owing c ompar i sons :

( 2 . 16) a. l i p : see ( 2 . 2 : n ) b. l i b : see ( 2 . 3 : n )

c. I It: see ( 2 . 4 : n ) d. l i d : see ( 2 . 5 : n)
e. I lk : see ( 2 . 6 : n ) f. I /g : see ( 2 . 7 : n )
g. I I s : see ( 2 . 8 : n ) h. I / h : see ( 2 . 9 : n )
i. I I e : s ee ( 2 . 1 0 : n ) j . I / J : see ( 2 . 11 : n )
k. 1 1m : see ( 2 . 1 2 : n ) l. I I n : see ( 2 . 1 3 : n )
m. I I fl : s e e ( 2 . 1 4 : n ) n. I I I) : s e e ( 2 . 1 5 : n )
o. I I r : I ta I i I [ ta I i J ' r op e ' - Itar i l [ ta r i J ' danc e '
p. I /w : I b il l a l [ b a l a J ' tr o o p ' - I b a w a l [ b aw a J ' carry '
q. I /y : I l a l ul [ I a I u J ' pa s t ' -
I l ayul [ l a y uJ ' w i t h e re d '

The under l y ing c on sonant I I I i s phonet ic a l l y r e a l i sed a s a vo i c e d

a l v e o l ar fr i c t ion l e s s lat eral s o und ( [ I J ) .

2 . 1 . 1 . 16 . The U n derl ying Con sonant Irl

T h i s und e r l y ing c o n s onant may o c c ur i n morpheme i n i t i a l , med i a l and

f i na l p o s it i on s . I t s ident i t y and d i st i n c t ive f un c t i o n c an b e e s t a b -
l i shed by t he fo l l ow i ng compar i son s :

(2. 17) a. r i p : see ( 2 . 2 : 0 ) b. r i b : see ( 2 . 3 : 0 )

c. r / t : see ( 2 . 4 : 0 ) d. r i d : see ( 2 . 5 : 0 )
e. r / k : see ( 2 . 6 : 0 ) f. r i g : see ( 2 . 7 : 0 )
g. r I s : see ( 2 . 8 : 0 ) h. r I h : see ( 2 . 9 : 0 )
i. r i c: s e e ( 2 . 1 0 : 0 ) j . r / J : see ( 2 . 11 : 0 )
k. r im : s e e ( 2 . 1 2 : 0 ) l. r I n : s e e ( 2 . 13 : 0 )
m. r / fl : s e e ( 2 . 1 4 : 0 ) n. r / l) : s e e ( 2 . 1 5 : 0 )
o. r I I : see ( 2 . 16 : 0 )
p. r / w : Ime r a hl [ me r a h J ' re d ' -
I m ew a h l [ m ewa h J ' l ux uri o u s '
q. r / y : I s a r a l) l [ s a r a l) J 'nes t ' - I s a y a l) l [ s a y a l) J ' to love '
 25

The under l y ing con sonant / r / i s b asic a l ly r e a l i sed on t he phone t i c

l ev e l a s a v o i c e d ap i c o a lveo lar t r i l l ( [ r J ) . The und e r l y ing / r / i n
t he pre f i x e s / b a r / ( ' stat i ve /a dj e c t i val verb p r e f i x ' ) , / t a r / ( ' adverb/
sup er lat ive forming pr e f ix ' ) and / p a r / ( ' c ausat ive forming p r e f i x ' ) is
p honet i c a l l y re al i s e d a s [ 0 J when t he prefix e s are adde d t o s t e m s
b eg inn ing w i t h / r / , suc h a s : /bar + rapa t/ [ barapa t J ' to h a v e m e e ti n g ' ,
/tar + ramay/ [ t a r ama y J ' th e m0 8 t a r owde d ( p "l a ae ) ' , /par + randah/
[ pa r an d a h J ' to "l ower ' , e t c . ; or when t he y are added t o a s t em who s e
f i r s t s y l l ab l e c on t a i n s a s c hwa f o l lowed b y a t r i l l s u c h a s / b a r +

ka r J a / [ b a ka r J a J ' to w or k ' , / ta r + p a r <:: a y a / [ t a p a r c a y a J ' th e m0 8 t

tr u 8 te d p er s on ' , / pa r + tarnak + ani [ pa ta rn a k a n J 'rana h ' , e t c . The
unde r l ying / r / in t he pr e fix / b a r / i s p honet i c a l l y re a l i se d as [ I J (a
v o i c e d a lveo lar l at e ral c o n sonant ) when t he pr e fi x / b a r / i s add ed t o
t he i t em ( st em ) / a J a r / ' to te a a h ' . Thus t he underl y i ng f orm / b a r +

a J a r/ ' to 8 tudy ' ( s emant i c a l l y i s t o t eac h one s e l f ) i s p honet i c a l l y

r ea l i se d a s [ b a l a J a r J . Thi s e x c e p t ional i t em ha s t o b e fu l l y sp ec i fi e d
in t he l e x i c o n . ( S ee a furt her d i sc u s s io n o f t h i s p henomenon in
Sect ion 5 . 1 . )

2 . 1 .1 . 17. The Underlying G l ides / w / and / y /

Bot h o f t he s e und e r l y ing g l id e s m a y o c c ur in morpheme i n it i a l ,

m e d i a l and final po s it ion s . We saw in t he c ompari sons g iven above t hat
e ac h of t hem c ont rast s w i t h t he 1 6 underl ying c on sonan t s di s c u s sed
above . There fore i t is s u f fic i ent to o p p o s e t hem t o eac h ot her t o s ho w
t ha t e a c h o f t hem ha s d i f ferent i dent i t y and d i st inct i ve func t i o n .
The s e t wo g l i d e s are c ontrast ive a s shown b y t he p a i r s : / r aw a / [ r a w a J
' 8wamp ' v s / r a y a / [ r a y a J ' "la r g e ' ; / b a w a l) / [ b a w a l) J ' on i on ' v s / b a y a l) /
[ b a y a l) J ' 8 ha d ow ' ; / r a n t aw / [ r a n t a w J ' ov e r 8 ea 8 ' v s / r a n t a y / [ r a n t a y J
' a ha i n ' , e t c .
The under lying g l ide / w / i s p ho net i c a l l y r e a l i s e d as a v o i c e d lab i o ­
v e l a r g l id e ( [ w J ) . The under l y ing g l ide / y / i s p honet i c a l ly rea l i se d
a s a v o i c e d l amino palat al g l i d e ( [ y J ( = I P A [j J ) ) .

2 .1. 2 . L O A N C O NS O NA NT S

A s ha s already been ind i c a t e d i n Chart 1 , t he r e are f o ur l oan con­

sonant soun d s whic h have been intro duc e d int o t he sound sy st em o f B I
t hrough loanwords . They are , a l l fri c at ive s , [f ] , [s ] , [ z J , and [ x J .
Eac h o f t h e s e sounds i s usua l l y in free variat ion with a cert a i n
' nat ive ' sound i n most l o anwords suc h as ( 2 . 18 ) .
26

(2. 18) 1. pabr i k [ pab r , ? , fab n ? ] ' man ufa c tuI'e '
hapa l [ ha pa l , ha fa l ] ' to m emor i s e '
ta rap [ t arap, taraf] ' p ha s e '

2. syah, sah [ sa h , sah] ' 'l ega 'l '

a sy i k , as i k [a s ' ? , aS l ? ] ' ab s orb e d '
m a s y h u r , m a s h u r [ ma s h o r , ma s ho r ] ' famou s '

3. zama n , j am a n [ J ama n , s am a n , zama n ] ' e ra, period '

za t [ sa t , za t ] ' s ub s ta n ce '
a s a s , a z a s C as a s , a za s ] ' pr i n c i p 'l e '

4. a k h i r , a k i r [ a k l r , a h , r , aX l r ] ' en d '
k h a ba r , k a b a r [ ka b a r , h a b a r , x a b a r ] 'news '
i kh l a s , i kl as [ i k l as , i h l as , ixl as] ' s i n ce r e '

5 . wa g o n , w a h o n [ w a g � n , w a h � n , w a x� n ] ' wagon '

a g e n [ a g en , a h e n , ax e n ] 'agent, agency '
b i o l og i [ b i o l o g i , b i o l o h i , b i o l o x i ] ' b i o 'l og y '

From t he above e xamp l e s we s e e t ha t [ f ] and [ p ] , [ 5] and [ 5 ] , [ z ] and

[ J ] or [ 5 ] , and [ x ] and [ k ] or [ g ] or [ h ] are i n free variat ion i n t he
s p e e c h c ommuni t y a s a who l e , and t o some ext ent , in t he spe e c h o f an
i n d i v i d u a l speaker in t he sense t hat he may u s e a loan sound w i t h c e r ­
t ai n l oanwor d s b u t s ub s t i t ut e s a nat ive sound f o r t he l o a n sound i n
o t her l o anwor ds . Thi s p henomenon o f fr ee var iab i l it y b et ween a fore ign
and a nat ive sound is not a b s o l ut e l y t ru e i n a l l c a s e s . There are
c er t ain l oan i t em s , p art i c u l a r l y items c o nt a in ing t he lab i a l f r ic at ive
( [ f ] ) r e c ent l y borrowed who s e for e i gn c har ac t er i s t ic s are pre s erved t o
s o me degree part l y b e c au s e t he y may b e i n c on f l i c t ( i .e. t he y appear t o
form m i n imal pa i r s ) w i t h it ems t hat a lready e x i s t ( c f . 2 . 1 9 ) and p a rt l y
b e c a u s e t he y a r e spe c i a l voc a b u l ar y i t em s u s e d b y e duc at e d p e o p l e who
hap p e n to be fam i l ia r w i t h t he sourc e language ( s ) (cf. 2 . 20) .

( 2 . 1 9 ) la . vak [ fa k] , s c hoo 'l s ubj e c t '

b. p a k [ pa k ] ' p a c k ag e '

2a . f a k t a [ fa k t a , pakta ] ' fa ct s '

b. p a k t a [ pa k t a ] 'pa c t '

3a . vas [ fa s ] ' va s e '

b. p a s [ pa s ] ' pr e c i s e , fi t '

4a. sa rap , sa raf [ s a r a f , sarap] ' ne r v e '

b. s a r a p [ sa ra p ] ' g a r bag e '

5a . z e n i [ zen i , sen i ] ' ( army ) e n g i n e e r s '

b. sen i [ san i ] 'art '
 27

6a . z a m z a m [ z am z a m , j a m j a m J ' h ol y wa ter '

b. j am j a m [ J a m j a m ] ' w a tc h e s '

7a. sya rat [ sara t , sarat] ' r e q ui r e me n ts '

b. sarat [ sa ratJ ' (fu l ly ) l oa d e d '

8a. k h a s [ x a s , ka s , h a s ] ' ty p ic a l '

b. ka s [ ka s ] ' safe , m on e y s upp l y '

Not i c e t hat eac h o f t he a - i t ems i n t he above examp l e s forms some sort

of minimal pair w i t h t he b - it ems in t he corr e s p ond ing number . The
e x i s t en c e of t ho se p a i r s ( al t hough t hey may s omet ime s app ear i d e nt i c a l
on t he p honet i c l eve l ) ha s b een made t he b a s i s for rec ogni s ing t he l o an
sound s a s c on s t it ut ing di fferent p honeme s . Not i c e , furt hermor e , t hat
[ f ] in t he fo l low i ng examp l e s ( 2 . 2 0 ) is har d l y e ver to my knowle dge
r e p l a c e d wit h [ p ] .

( 2 . 20) f a s i l i t a s [ fa s i l i t a s ] ' fa c i l i ti e s '

federa s i [ fa d a r a s i ] ' fe de r a ti on '
f a ku l t a s [ f a ku l ta s ] ' fa cu l ty '
f a k to r [ f a k t � r 1 ' fac tor '
f a na t i k [ fa n a t i ? ] ' fana ti c '
va r i as i [far i as i ] ' v aria ti on '
v e to [ fe to ] ' v e to '
v i sa [ f i s a ] ' vi sa '
v i ta l [ f i t a l ] ' v i ta Z '
va l u ta [ fa l u ta ] ' c urrenc y '

Examp l e s ( 2 . 1 8 ) sugg e st t hat we do not need t o p o s t u lat e t he loan

c on s onant s a s under l y i ng s e gmen t s b e c a u s e we c an alway s c h o o s e one of
t he nat ura l i sed ( p honet ic ) forms a s t he und er l y ing repre s ent at i o n for
e a c h o f t ho s e l oan i t ems and a c c ount for t he fore i gn ( p ho ne t i c ) a l t e r ­
nant s by rul e s . T h i s proc e dure , alt hough i t save s a f e w symb o l s , has
some apparent d i sadvant age s . F i r s t l y , we c annot make a real general
s t at ement about t he real i sat i o n o f t ho s e l oan i t ems b e c au s e s ome of t he
l oan i t em s require t he o b l igatory app l i cat ion of phone t ic re a l i sat i on
rul e s , some l oan items require t he opt ional app l i c at i on of c ert a i n
r e a l i sat ion r u l e s and some l o a n items do n o t require t he app l i cat i o n o f
c er t a i n rea l i sat i o n r u l e s a t a l l . For examp l e , if we take / p / as t he
under l y ing form o f [ f ] , t he n we n e e d a p honet i c r e a l i sat i o n rule t o
c onvert t he p lo s ive into a fr i c at i ve . Thi s rule app l i e s opt i o na l l y t o
i t e m s l ike t ho s e g i ven i n ( 2 . 18 : 1 ; 2 . 1 9 : 2 a , 4 a ) , i t app l i e s o b l i gat or i ly
t o i t em s l ike t ho se g iven in ( 2 . 2 0 ) , but it doe s not app ly at a l l t o
i t e m s l ike t ho se given i n ( 2 . 1 9 : 1 b , 2b , 3b ) a l t hough they are a l s o l oan
items . The r e fore we have to mark i n t he l e x i c o n whet her a c er t a i n l oan
28

i t em h a s t o undergo , may or may no t undergo a c ert ain p ho net i c rea l i sa­

t ion rul e .
S ec ondly , we wi l l not always in su c h c a se s b e able to have di f ferent
underly ing repre sentat i o n s for d i f ferent morpheme s ( c f . 2 . 19 ) . Con­
s equent l y , forms l i ke t ho se in ( 2 . 1 9 ) have t o b e i n d e x e d a n d marked
along w i t h t he indi c e s whet her t hey must , or may , or may not undergo
a c er t a in p honet i c rea l i s at ion rul e . Obviou s l y , t h i s s aving of s ymb o l s
d o e s not make t he grammar s imple . O n t he c o nt rary , it c omp l i c at e s and
ob s c ur e s t he l e x i c on .
Now we are l e f t w i t h t he s e c ond a l t ernat ive nam e l y t o p o s t u l a t e t ho se
l oan c o n sonant s a s unde r l y in g segment s in addit ion t o t he 1 8 con sonant s
o c c urr ing in nat ive BI i t ems d i s c u s sed in t he previous sub - s e c t ion .
The advant age o f t hi s proc edure i s obviou s . We c an t hen genera l i se
t he pro c e s s o f t he subs t i t ut i on o f t he loan sound s w i t hout c omp l i cat ing
and o b s c ur i ng too much t he l ex ic on part o f t he grammar .
The p o s t u lat ion o f t he four fore i gn sounds as under lying c on sonant s
ha s not however so lved al l pro b l em s c on c erning loan items . The fa c t
t hat c er t a in items m a y no t undergo c er t a in r e a l i s at ion rul e s ( e . g . it ems
l i ke t ho s e g iven in ( 2 . 2 0 ) whi c h are no t l ia b l e t o t he rule t hat c onvert s
a l a b i a l fr i c at ive int o t he c orre spond ing stop ) make s it impo s s ib l e t o
avo i d t he use o f d iacrit ic feat ure s ( se e a l s o S e c t ion 5 . 4 . ) .

2. 2. V OW E L S

Mo st wor k s d e a l i ng wit h t he sound s y st em o f B I c la im t hat t here are

s ix vowe l p ho n em e s in B I ( se e , e . g . Al i sj ahb ana 1 95 1 : 1 2 - 1 5 ; Halim 1 9 7 4 :
1 69 - 1 7 3 ; a l so H a l im 1 97 2 : 1 5 4 ; Johns 1 9 7 5 : 2 - 7 ; Kahler 1 9 5 6 : 3 4 ; Mac donald
and Dardj owidj oj o 1 96 7 : 2 1 - 3 0 ; Panitia Edj aan Lembaga Bahasa dan
l
Kesusasteraan 1 9 6 6 : 8 ; Wo l f f 1 9 6 5 : 5 , et c . ) . The s i x vow e l phonemes are
u s ua l l y repr e s ent e d by t he symbo l s : Ii , e, a, a, ul and 101 . These
vowe l s can furt her b e c haract e r i sed in general phone t ic t e rms b y means
of t he f o l l owing c hart :

l
Halim ( 1914 : 169ff ) report s that Samsuri ( 1960) and Dardj owidj oj o ( 1966 ) , who are
both Javanese speakers , posit eight vowels for BI under the influence of Javanese
which recognises eight vowel phonemes , i . e . three front vowels ( / i , I , e/ ) , two c entral
vowels ( /a , a/ ) and three back vowels ( /u , o , o/ ) . It will be clear later that [ I J and
[oJ are respect ively the non-tense versions of I i i and 101 ( see Sect ion 5 . 6 . ) .
 29

CHART 2

Unrounded Rounded

Front C entral Bac k

High i u

Mid e a 0

Low a

The above s i x vowe l s are a l l p o s t u lat e d as underl y i n g vowel s egment s

in t h e pr e s ent d e s c ript ion . In t he sub s e quent paragraphs we are go i n g
t o e xamine t he r e l at i o n s h i p s b et ween t he under l y ing vowe l s w i t h regard
t o each ot her ( i . e . t he i r d i s t i nc t ive r e l at ion ) and b et w e e n t he under­
l y i n g vowe l s and t h e ir r e s p e c t i ve p honet ic r e a l i sat ion s . S in c e t he
vowe l s l e i and l a l never form minimal p air s ( at least not w i t hin t he
range o f nat ive BI i t ems ) in whic h t he y can b e c ontra st e d , t he d i s ­
c u s s i o n o f t he s e t wo under l y ing vowe l s w i l l c ome last . The order o f
t he d i s c u s s ion o f t he under l y i ng vowe l s b e low ha s no s i gni f i c anc e what ­
soever . It w i l l be s e e n from t he c ompa r i s on s g i ven b e low t hat l a l i s
never stre s s e d . T h i s mean s t hat i t ems c on t a i n ing l a l w i l l not c onst i ­
t u t e a b s o l u t e minimal p a i r s wit h i t ems c ont aining ot her vowe l s . The
fac t s t ha t high vowe l s I i i an d l u i may b ec ome non - s y l la b i c ( c f . S e c t i on
5 . 7 . ) make s it nec e s s ary to c ontrast t he high vow e l s w i t h t he c orre s­
p onding g l i de s I y l and Iwl r e s p e c t ive l y .

2.2 . 1 . T H E U N D E R L Y I NG VOWEL Iii

T h i s und er l yi ng v ow e l may oc c ur in morpheme i n i t i a l , me d i a l and

final p o s i t ion s . I t s ident it y and d i s t inct ive func t ion c an be e st ab ­
l i shed b y t he fo l lowing compar i son s :

( 2 . 21 ) a . i / a : I i ka n l [ i k a n ] ' fi s h ' -
l a k a n l [ a ka n ] 'wi l l '
b. i l u : I g i I a l [g i I a] 'mad ' -
I g u l al [ g u l a ] ' s ugar '
c. i /o : I k i tal [ k i t a ] ' we ( inc 1 . ) , -
I ko t a l [ ko .t a ] , town, c i t y '
d. i I e : Ib i I al [b i I a] ' wh e n ' -
I be l al [ be l a ] ' de fe n d '
e. i /a : I k i ral [ k i r a ] ' think ' -
I ka r a l [ k a r a ] 'mo n k e y '
f. i Iy : l i al [ i a] ' he , she ' -
Iya I [ y a ] 'yes '

T he und e r l y ing vowel I i i i s b a s i c a l l y r e a l i se d as a voic e d high

front unround e d oral vowe l ( [ i ] ) on t he p hone t i c l e ve l . I n final c lo s e d
syllable ( i . e . s y l l ab l e ending i n a c on sonant ) , t he under lying vow e l I i i
i s p ho ne t i c a l l y real i se d a s a non-t ense high front unround e d vowel ( [ i ] )
(cf. S ec t i o n 5 . 6 . ) . Somet ime s t he und er l y ing I i i may a l s o b e r e a l i s e d
30

p honet i c a l l y a s a vo i c ed p a l a t a l gl i de ( [ y ] ) in c er t a i n env ironment s

such a s I s u r i a l [ s u r i a , surya] ' t he sun ' , I i u r a n l [ i u r a n , yuran] 'con­
t r i b u t i o n (money ) ' , e t c . ( c f . S e c t ion 5 . 7 . ) .

2 . 2 . 2. T H E U NV E R L Y I NG V O W E L lal

T h i s under ly ing vowe l may oc c ur i n morp heme i n it ia l , medial and

final p o s i t ion s . I t s ident i t y and d i st inct ive func t ion can be e st ab ­
l i shed b y t he f o l l owing compar i so n s :

( 2 . 22 ) a . a l i : see ( 2 . 2l : a )
b. a/ u : I b a t a l [ ba t a ] ' b r ic k ' -
Ibatul [batu] ' s tone '
c. a / o : I ka t a I [ k a t a ] ' wor d ' -
I ko t a l [ k o t a ] , town, city '
d. ale : lanakl [ana?] ' ch i l d ' -
l e n a kl [ e na ? ] ' de l i c i o u s '
e. ala : I s a r a f) 1 [ s a r a f) ] 'nest ' -
I s a r a f) 1 [ s a r a f) ] ' a t ta c k '

The under lying vowe l l a l i s p hone t i c a l l y r e a l i s e d a s a vo ic e d c ent ral

l o w unr ounded oral vowe l ( [ a ] ) .

2. 2 . 3 . T H E UNV E R L Y I NG V O W E L l ui

T h i s underlying vowel may o c c ur in morp heme init ial , me dial and

final po s i t i o n s . I t s ident i t y and d i st inct ive funct ion c an b e e s t a b ­
l i s hed b y t he f o l l owing c ompar i sons :

(2. 23) a. ul i : see ( 2 . 2 l : b )

b. u/a : see ( 2 . 2 2 : b )
c. u/o : I p u l al [ p u l a ] 'also ' -
I po l a l [ po l a ] ' p a t tern '
d. u/e : I t undal [ t unda ] ' p o s tpone ' -
I tendal [ tenda] ' tent '
e. u / a : l u l a f) 1 [ u l a f) ] ' re p e a t ' -
l a l a f) 1 [ a l a f) ] ' hawk '
f. u / w : I ka u 1 I [ k a o l ] ' promi s e ' -
I ka w u l l [ k a wo l ] ' t i n der '

The und erly ing l u i i s b a s i c a l ly rea l i se d as a v o i c ed high b a c k

rounded o r a l vowel ( [ u ] ) o n t he p honet ic leve l . In a f i n a l c lo s e d
s y l l ab l e , t he under ly ing l ui i s p honet i c al l y rea l i s e d a s [ o ] , a lax
version of [ u ] ( c f . Sect ion 5 . 6 . ) . Somet ime s t he under l y i ng l u i i s
al s o rea l i s e d a s [ w ] i n c er t a i n environment s suc h a s l u a f) 1 [ u a f) , w a f) ]
' mo n e y ' , I k u i t a n s i l [ k u i t a n s i , kw i t a n s i ] ' rec e ip t ' , e tc . ( c f . Sect ion
5 . 7. ) .

2. 2. 4 . T H E U NV ER L Y I NG VOWEL 101

Th i s und er l y i ng vowel may o c c ur in morpheme init ial , medial and

,
f i na l p o s i t i o n s .
I t s ident it y and d i st in c t ive func t ion c an be e s t ab ­
l i shed b y t he fo l lo wing c ompar i sons :
 31

( 2 . 24 ) a . ol i : see ( 2 . 21 : c l
b. o/ a : see ( 2 . 2 2 : c )
c. o / u : see C2 . 2 3 : c )
d. o l e : I bo l a l [ b o l a J ' ba l l ' - I be l al [ be l a] ' defend '
e. o/a : I rodal [ rodaJ 'whee l ' - I r a d a l [ r ad a J ' s top ( r ai n ) ,

The under lying 1 0 1 i s b a s i c a l l y real i sed a s a v o i c e d m i d bac k

rounde d oral vowel ( [ o J ) on t he p ho net ic level . In a f i n a l c l o s e d
s y l l ab l e , t he under l ying 1 0 1 i s p honet i c al l y r e a l i se d a s [ o J , a l ax
vers ion of [ o J ( c f . S e c t ion 5 . 6 . ) .

2.2.5. T H E U ND E R L Y I NG VOWE LS / e / AND / a /

The und e r l y i ng vowel l e i may o c c ur i n morp heme i n i t i a l , med i al and

f i n a l p o s it io n s . I t i s b a s i c a l l y r e a l i s e d as a vo ic e d mid front un­
rounded oral vowe l ( [ e J ) on t he phone t i c l evel . In a f i n a l c lo se d
s y l l ab l e , t he under l y i n g l e i i s p honet i ca l ly r e a l i s e d a s [ E J , a l a x
vers ion o f [ e J ( c f . S e c t ion 5 . 6 . ) .
The und e r l y ing vow e l l a l may o c c u r i n morpheme i n i t i al and med i a l
p o s i t ions only i n nat iv e B I ( Mal a y ) i t ems . It i s p honet i c al l y r e a l i s e d
a s a vo i c ed c entral m i d oral vowe l ( [ a J ) . S o met imes t he under ly ing l a l
i s phone t i c a l ly real i se d a s [ 0 J i n c e r t a i n environment s such a s l a m a s l
[ am a s , m a s J ' g o l d ' , I k a r a n a l [ ka r a n a , ka r n a J ' b e ca u s e ' , e t c . (cf.
S e c t ion 5 . 6 . ) .
The fact t hat l e i c an b e real i s e d a s r a J on t he p honet ic l e v e l r e ­
qu i r e s some j ust i f i c at i o n f o r t he p o s t u l at i o n o f l a l . Phono l og i c al l y ,
l a l b ehav e s d i f f e rent l y from t he o t he r f ive vow e l s i n some r e s p e c t s .
F i r st l y , t he sc hwa doe s not oc cur in morpheme f i na l syl lab l e e x c e p t in
a small numbe r of r e c ent l y borrowe d i t ems suc h as ( 2 . 2 5 ) :

( 2 . 25) r uw e t ( Javane s e ) [ r uw a t J ' comp l i ca t e d '

g a n t e n g ( Javan e s e ) [ g a n t a f) J ' handsome '
b a n t e r ( Javane s e ) [ b a n t a r J 'quick '
t a n t e ( Dutc h ) [ t a n t a J ' a un t '
s i s t e m ( Eng l i s h/Dut c h ) [ s i s t am J ' sy s t e m '

Many speaker s however t end t o u s e [ e J ( or [ E J - the lax ver s i on o f [ e J )

i n s t e ad , part l y b e c a u s e nat ive Mal a y ( B I ) it ems do not c on t a i n s c hwa
i n t he f in a l s y l l a b l e and part ly b e cause many r e g ional l anguage s do not
r e cogn i se t he sc hwa at a l l .
S e c ond l y , t he s c hwa i s u s u a l l y unstre s s ab le w h i l e t he o t her vowe l s
are p o t ent i a l l y stre s sab le . I n gener a l , st re s s f al l s o n t he penult ima t e
syl l a b l e o f wor ds . But when t he vowel o f t he penult ima t e s y l la b l e i s
a s c hwa , t h e st re s s i s s h i ft e d t o t he ult ima i n b i sy l l a b i c wor d s and
32

a l s o i n p o l y s y l l a b i c words when t he s y l lab l e s b efore t he p e nu l t a l s o

c on t a i n [ a ] a s in ( 2 . 2 6 : 1 ) , or t o t he ant e penul t imat e syl l ab l e in p o ly ­
s y l l a b i c word s provided t hat t he vow e l o f t he ant e p enult i s not a s c hwa
a s in ( 2 . 2 6 : 2 ) b e l ow . ( See a furt her d i s c u s sion o f thi s phenomenon in
Sec t io n 5 . 9 . ) .

( 2 . 2 6 ) la . e m a s C am a s ] 'gold ' b. seteru [sataru] ' enemy '

terang [ taraQ] ' light ' tente ra [ tantara] ' s o l di e r '
sed i h [ s ad rh] ' sa d ' s edekah [ s adakah] ' a lm '
b e r i [ ba d ] 'give ' ke l ed a i [ k a l a da y ] ' don k e y '
penuh [ panuh] ' fu l l ' s e l esa i [ s a l asay] ' fi n i s he d '

2. c e r i t e r a [ c a r r t a ra , c a r r t ra ] ' s tory '

p u t e r a [ pu t a ra , p u t r a ] ' so n , boy '
s u tera [ s u ta r a , s u t r a ] ' si l k '
s am u d e r a [ s am u d a r a , s a m ud r a ] ' o cean '
m a n t e r a [ ma n t a r a , m a n t r a ] ' ma g i c wor ds '

No t ic e t hat [ a ] in e xamp le s ( 2 . 2 6 : 2 ) above c an be d e l e t ed on t he phon­

et i c l e ve l . I n fac t , sc hwa t end s t o be d e l e t e d in c a su a l s p e e c h . ( Se e
a furt her d i sc u s s i on o f t hi s s c hwa de l e t i o n in S e c t ion 5 . 6 . )
A p r o b l em ari s e s when we c ompare t he p hono l og i c a l b e haviour o f [ a ]
w i t h t ha t o f [ e ] . The front mid vowel ( [ e ] ) u s ua l ly o c c ur s o n l y in t he
penult and ult ima in p o l y sy l l ab i c i t em s . The oc c urrence o f [ e ] i n
s y l l ab l e s o t h e r t han t he penult o r u l t ima in p o l y s y l lab i c i t em s ( i . e .
i n u n st r e s s ed p o s i t ion ) i s limited t o a numb e r o f l oan i t ems s u c h a s
( 2 . 27 ) :

( 2. 27) D e s em b e r [ d e s em b e r ] ' Dec ember '

repub l i k [ repub l i k] ' re p ub l i c '
d em o k r a s i [ d em o k r a s i ] ' demo crac y '
s e r t i p i ka t [ s e r t i p i ka t ] ' ce r tifica te '

It mu st b e no t e d however t hat many speaker s t end t o r e a l i s e t he un­

s t r e s s e d mid front vowe l in t he above e x amp l e s as [ a ] on t he phone t i c
l eve l . Thi s t endency i s al so o b served i n wor d s derived from s t ems con­
t a in ing t he front mid vowel a s in ( 2 . 2 8 ) :

( 2 . 2 8 ) la . bebas [ be b a s ] ' fr e e '

b. k e b e b a s a n [ ka b a b a s a n ] ' fr e e dom '

2 a . me r de k a [ m a r d e k a ] ' i n de p enden t '

b. k em e r d e k a a n [ k a m a r d a k a 7 a n ] ' in de p e n denc e '

3a . desa [desa] ' v i l lage '

b. p ed e s a a n [ p a d a s a 7 a n ] ' v i l lage area s '
 33

No t e t hat t he a - i t ems in t he above e x amp l e s are st ems and t he b - i t ems

are derivat ive s . Not e , furt hermor e , t ha t st re s s i n t he ab ove deriva­
t iv e s is shift e d w i t h r e s p e c t to t he st re s s a s signed t o t he s t ems .
Up t o t hi s p oint we might wi s h t o regard t he s c hwa ( [ a J ) s imp ly a s
a n al lophonic rea l i s at ion o f l e i b e c au se i t s d i st r ibut ion s eems t o b e
p hono logic a l l y definab l e , i . e . it o c curs in a non-final un s t re s s e d
p o sit ion , a n env ironment wher e [ e J doe s n o t oc c ur . ( It ems c i t e d in
( 2 . 2 5 ) and ( 2 . 2 7 ) can b e ac count e d for b y a s s igning t he diacr i t i c
f eat ur e loan in t h e l e x i c o n . ) Thi s kind o f s o l u t i on apparent ly i nt ro ­
duc e s a new pro b l em . Unl e s s a c o n straint or a se t o f const raint s e x i s t
wh ic h s ho w t hat t he t wo vowel sounds ( [ e J and [ a J ) a r e mut ua l l y e x c l u ­
s ive in t erms o f vowe l s t hat m a y p r e c ede or fo l low e a c h o f t hem , stre s s
mu st b e h e l d r e s p o n s i b l e f o r t he d i st r ibut i o n o f t ho s e vowe l s . In o t her
wo rd s , s t r e s s ha s to b e r egar ded a s hav ing d i st inct ive func t io n in t he
s e n s e t hat it i s l e x i c a l l y idi osyncrat i c and t he r e fore unpr e d ic t ab l e .
Examp l e s ( 2 . 2 9 ) be low may suc c in c t ly demon s t ra t e t hat b y and large
str e s s is predic t ab l e in B I . In gener a l , one c an s ay t ha t st r e s s in
B I fal l s on t he p enult ima t e s y l lab l e o f wo r d s . Thi s i s c learly shown
b y i t ems b and c b e l o w .

( 2 . 2 9 ) la . b o d o h [ b6 d o h J ' s tup i d '

b. k e b o d o h a n [ k a bo d 6 h a n J ' s tup i di t y '
c. k e b o d o h a nm u [ k a bo d o h a n m u J ' your s t upi di t y '

2a . pendek [ p�ndE?J ' s hor t '

b. mem e n d e k k a n [ ma m a n d e k a n J ' to s ho r t e n '
c . m em e n d e k k a n n y a [ m am a n d a k a n p a J ' to s hor t e n i t '

3a. be l a [ be l a J ' to defe n d '

b. p em b e l a a n [ p a m b a l a ? a n J ' de fe n a e , a a t of de fe n din g '
c. p em b e l a a n k u [ p a m b a l a ? a n k u J 'my de fe n ae '

4a . t e r a n g [ t a r a !) J ' light, bright '

b. d i t e r a n g k a n [ d i t a r a !) k a n J ' t o b e exp l a i n e d '
c. d i t e r a n g k a n n y a [ d i t a r a !) k a n p a J ' to be exp l ai n e d by him '

5a . ber i [badJ ' to give '

b . p em b e r i a n [ p a m o a r f a n J ' g i ft '
c. p e m b e r i a nm u [ p am b a d a nm u J ' your gift '

6a . cer i tera [ca r f taraJ ' s t or y '

b. c e r i t e r a k a n [ ca r i tarakanJ ' t o t e ll. a s tory '
c. d i cer i ter akannya [ d i c ar i tarakanpa] ' i t w a s t o l d by h i m '

Alt ho ugh t he p enult ima t e s y l lab le of t he l a s t t hree st ems ( 2 . 2 9 : 4 a , 5a ,

6a 1 i s not stre s s e d , t he p l a c ement o f st re s s in words derive d from t he
s t em s ( c f . 2 . 2 9 : 4 b , c ; 5b , c ; 6b , c ) i s o n t he penult imat e s y l l a b l e o f t he
der ivat i ve s . Ther e f or e , t here are good rea s o n s for not r e garding st re s s
a p honeme i n B I . T h i s mean s t hat t he d i s t r ibut i o n s o f [ e J and [ a J
c annot be regard ed as b e ing c ondit i oned b y s t re s s , but r at her t he o t her
l
way round .
I n or der to j u st i f y t he int erpr e t at ion o f s c hwa as t he a l l ophone o f
l e i , w e mu st t here fore r e l y on t he e x i s,t enc e o f a c o n st raint or a s e t
o f c on s t ra int s ( i . e . morpheme struc t ur e c9ndit i on s ) whi c h c an b e he ld
r e s p o n s i b l e for t he d i st r ibut ions of t he se vowe l s ( I e J and [ a J ) . I t has
b ee n p o inted out earl ier t hat [ e J and [ a J are mut ua l l y e x c l u s ive i n
morpheme final s y l lab l e , a n d t o a degr e e in morp heme i n i t i a l a n d med i a l
s y l l ab le s ( sy l lab l e s o t her t ha n t he p enul t ) i n p o l y s y l l a b i c i t ems . In
t he p enult ima t e s y l l a b l e , t he t w o vowe l s ( [ e J and I a J ) a r e a l s o i n c om­
p l ement ary d i st r i but ion i n c er t a in environment s in ind igenou s i t ems .
The vowe l [ e J may o c c ur in t he p enult imate syl lab le when t he ult ima
c ont a i n s a m id vowe l as i n ( 2 . 3 0 : 1 ) , and t he vowel [ a J when t he ult ima
c ont a i n s a high vowel as in ( 2 . 3 0 : 2 ) .

(2. 30) 1. besok [ be s o ? J ' t omorrow ' pendek [ pend e ? J ' s hor t '
be l ok [ be 1 o ? ] ' t o t urn ' b e n t e ng [ b e n t e Q J 'fo r t '
e l ok [ e l o? J ' n i ce ' l ehe r [ l ehe r J ' n e ck '
t empo [ t empoJ ' t im e ' nenek [ nene ? J ' g randmo t h e r '

2. b e l u m [ b a l om J 'not y e t ' sed i h [ sa d i hJ ' sad '

t e r u s [ t a ro s J ' c o n t i nue ' p en t i ng [ p an t L Q J ' i mp or tan t '
seteru [ sataruJ ' e n emy ' l eb i h [ l ab i hJ 'more '
c e n d e r u ng [ ca n d a r o Q J ' tend ' ser i ng [ sa r L QJ ' often '

Ho weve r , b o t h vow e l s ( [ e J and [ a J ) may oc cur in t he penult ima t e s y l lab l e

when t he ult ima c on t a i n s a l o w vow e l ( [ a J ) suc h as ( 2 . 3 1 ) .

( 2 . 31 ) a . be l a [ be l a J ' d e fen d ' b. be l a h [ ba l a h J ' c ut L e n g t hw i s e '

re l a [ r e l a J ' w i L Ling ' reda [ radaJ ' s t o p ( r a in ) '
enak [ena?J ' de L i c i o u s ' e n am [ a n am J ' s ix '
bendera [ banderaJ ' fLag ' s e te l a h [ sa t a l a h J ' a ft e r '

l
The fact that stressed vowels are relat ively longer compared with unstressed vowels
might make us postulate an addit ional vowel in the underlying representat ion of items
whi c h are stressed on the final syllable on the phonetic level by doubling the stressed
vowe l . For example, the item terang [ taraQJ ' Light ' , b e r i [ ba r i J 'to give ' , etc . ( cf .
2 . 29 : 4a , 5a ) will have the underlying forms I te raaQI and I be r i i l respectively. This
would then account for the occurrence of IaJ rather than [ eJ in the first syllable of
the items (assuming that they are both the alternant s of le/ l bec ause the stress could
be viewed as occurring on the penult rather than on the ultima . To yield the correct
phonet ic forms of those items we would need a rule that would convert the two-ident ical
vowel sequenc es into single vowels . Thi s sort of rule does not seem to be phonologic­
ally motivated . As we shall see later , a sequence of two identical vowels i s usually
broken up by a glottal stop insertion rather than by degeminat ion .
 35

Wit hout t a king int o a c c ou nt min imal p air s whic h ex i st t hrough borrowing ,
s u c h as p e n i n g [ p an L Q] ' di z zy ' (BI ) vs p e n i n g [ p e n L Q ] ' a n ima Z Z i ce n ce
b a tc h ' ( Dutc h ) , teras [ t aras] ' p i th ' (BI) vs t e r a s [ te r a s ] ' te r r a ce '
( Dut c h/Engl i sh ) , et c . , e x amp l e s ( 2 . 3 1 ) above suc c in c t l y show t ha t t he
two vowe l s ( [ e ] and [ a ] ) have t o b e p o s t u l a t ed as t wo d i f fer ent under­
l y ing vowe l s i n t he phono log i c a l repre sent at io n .

2. 3 . T H E C HA R A C T E R I S AT I O N O F S E GM E N T S I N T E RM S O F D I ST I N C T I V E F E A T U R E S

I n generat ive phono l ogy , t he d e s c ript i on o n t he phono l o g ic a l and

p honet i c l e ve l s as w e l l as t he r u l e s t ha t c onvert p hono l o g i c a l repre s ­
ent a t i o n s int o p honet ic repre sent at i o n s are i n t erms o f d i s t i n c t ive
featur e s . The d i st inc t iv e feature s are " t he m i n imal e l ement s o f whi c h
p honet i c , l e x ic a l , and p hono log i c a l t ra n s c r ipt i o n s are comp os e d , b y
c om b i nat i o n and c onc at enat i o n " ( C homsky and H a l l e 1 9 6 8 : 6 4 ) . Thu s t he
f undam e n t a l unit o f a p ho no lo g i c a l d e sc r ipt ion i s the d i st i n c t ive
f e a t ur e . The phoneme ( de fine d as a b undle o f d i st i nc t ive feature s ) has
n o i nd e pend ent l ingu i st ic s t at u s ( se e e . g . Chomsky 19 : 6 9 ; F i s c her­
J�rgensen 1 9 7 5 : 21 8 , et c . ) . The a l p ha b e t ic symb o l s whi c h are u s e d t o
r e pr e s e nt sound segment s ( phoneme s o r a l l o p hone s ) are r e g arded a s n o t h i ng
more t han p ra c t i c a l ad hoc a b b rev iat i o n s for feat ur e bund l e s ( c f .
C homsky and Hal l e 1 9 6 8 : 6 4 ; Hal l e ( 1 9 6 2 ) 1 9 6 4 : 3 3 6 , et c . ) .
Hal l e « 1 9 6 2 ) 1 9 6 4 : 3 3 5 - 3 3 8 ) ha s c onvinc ingly shown t hat r ul e s are
frequent l y formula t e d muc h b e t t e r i n t erms of d i st in c t ive f e at ur e s t o
t he ext ent t hat general r ul e s c an b e e xpr e s sed mor e s imp l y t han l e s s
genera l r ul e s . T h i s i s not t he c a s e w i t h rule s formulat ed in t erms o f
a l p habe t i c symb o l s . The p ro b l em i s t o s e l e c t a n appropr ia t e s e t o f
f ea t ur e s from t he c ur r e n tl y ava i l a b l e f e a t ur e s whic h w i l l b e c apab l e
o f d e s c r i b ing t he s y s t emat ic p hone t i c s , o f d i f ferent iat ing l e x i c a l i t em s ,
and o f d e f i n ing natur a l c la s se s , t ha t i s , t ho s e segment s whi c h a s a
gr o up undergo s im i l ar phono l o g i c a l proc e s s e s ( c f . S c hane 1 9 7 3 : 2 5 , 3 5 ) .
In t he pr esent d e s c r ipt i o n , t he f e a t ur e s c ho sen are b a s i c a l l y from
t ho s e p r o p o s ed by Chom s ky and Hal l e ( 1 9 6 8 ) . They pr o p o s e 22 f e a t ur e s
whi c h t hey c la im t o b e univer sal and c an b e u s e d a s a framework for
d e sc r i b ing sound segment s o f any l anguage . For B I o n l y 1 3 feat ur e s ar e
u s ed t o char a c t e r i s e t he sound segm ent s d i s c u s s ed in t he l a st t wo s e c ­
t ion s . The 1 3 featur e s c o n s i st o f ( 1 ) t hr e e maj or c la s s featur e s
( c on sonant a l , sy l l ab i c , sonorant ) , ( 2 ) seven c av i t y fe a t ur e s ( c orona l ,
ant e r i or , high , low , b a c k , r ound , nasal ) , ( 3 ) t wo manner o f art i c ul a t i o n
f e a t ur e s ( c ont inuant , t en se ) , a n d ( 4 ) o n e source feat ur e ( vo i c e ) . Of
t.he 1 3 featur e s , o n l y 1 2 hav e c l a s s i fi c at o ry func t io n . O n e fea t ur e
(i.e. tens e ) i s r e l evant only t o t he d e sc r ipt ion ( repre sentat i o n ) o n
t he p ho net i c l eve l .
36

The f e a t ure syst em d e s cr i b e d i n T h e S o u n d Patt e�n 0 6 E ng i�� h ( Chomsky

and Hal l e 1 9 6 8 ) ha s been c r it i c i s ed o n v a r i o u s point s and t he r e fore i t
c annot b e r egarded as d e f i n i t ive ( s ee e . g . Lade foged 1 9 7 1 , Fant 1 9 7 3 ,
et c . ) . I n t he pr e s ent de s c r ipt ion , t h e segment s [ h J and [ 1 J are c l a s s i ­
f i ed , f o l lo w i ng Lad e fog ed ( 1 9 7 1 ) , a s non-sonorant g l i d e s ( [ -Con sonant ,
- S y l l ab i c , - Sonorant J ) . ( No t e t hat Chom sky and H a l l e t r e a t e d t he se t wo
segment s a s sonorant g l i d e s . ) The fe a t ur e syl labic , de s c r i b e d b y
C homsky a n d H a l l e ( 1 9 68 : 3 5 4 ) a s a feat ure " wh i c h wou ld c harac t er i z e
a l l segm e nt s c on s t i t ut i ng a s y l l a b l e p ea k" , i s u s e d her e in p l a c e o f
t h e f ea t ur e voca l i c o f t he i r featur e s y st em .
The f ea t ur e continuant in Chom s ky and Hal l e ' s fea t ur e s y s t em d i vide s
s o un d s into t wo group s on t he b a s i s o f t h e b l o c kage o f a i r f l o w in t he
vowel tract . They wr i t e " In t he produc t ion of c ont inuant sound s , t he
p r imary con st r ic t i o n in the vowe l t r a c t i s not narrowed t o t he po int
wher e the air flow p a st t he c on s t r i c t ion is b l o c ke d ; i n st op s t he a ir
f l o w t hr ough t he mout h i s e ffec t ively b l oc ked . " A c c ording t o t hi s
d e f i n i t i on , l iq u i d s ( lat eral and t r i l l ) are c ont inuant sounds b e c ause
t he air flow t hrough t he mout h is no t e f fe c t i ve l y b lo c k e d i n t he pr o ­
duc t i o n o f t he sound s . T o keep t hem apar t o n e need s t he f eature l a t er ­
ality . From t he s im p l ic i t y point o f v i e w , t h i s lat er a l i t y feature i s
t oo c o st ly for t he phonology o f B I b e c au s e i t ha s no p urpo se o t her t han
t o keep I I I and I r l d i st inct . For t h i s rea son , the tr i l l l i qu i d ( / r / )
i s c l a s s i f ied , f o l l owing Jakob son , Fant and Ha l l e ( 1 96 2 ) , a s a non­
l
c ont inuant ( or int e rrup t e d ) c o n sonant .
Tab l e 1 b e low s hows t he d i st in c t ive feat ur e c ompo s i t i on o f BI segment s .
The t a b l e ( or mat r i x ) i n d i c at e s whi c h s e t of featur e v a l u e s t he a lp ha ­
b et ic symbo l s u s ed i n t h e informal r epr e sent at i o n may b e re ferred t o .
F e a t ur e s whi c h are not r e levant t o c e rt ai n segment s are l e ft b lank .
Not i c e t ha t t he f ea t ur e tense and s egment s [ 1 J , [LJ, [EJ, [ o J and [ o J
are enc l o sed b e t ween bracket s t o i n d i c a t e t hat t he y have no c la s s i f i c a ­
t ory fun c t ion and t he r e fore t he y a r e n o t r e l evant t o t he d e s c r i p t i o n
o n t he phono l o g i c a l level .

l
In Jakobson and Halle ( 1956: 41-42 ) the bi nary oppo sition interrupted vs continuant
is replaced by abrupt VB continuant .
 37

TABLE

p b m t d n r 1 5 J'l .c J k 9 I) ( 7 ) Y w f z � x i (L) e (e:) a a U (0) 0 (0)

Consonantal + + + + + + + + + + + + + + + - + + + + -
Syllabic - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + +
Sonorant - - + - - + + + - + - - - - + - + + - - - - + + + + + + + + + +
Coronal - - - + + + + + + - - - - - - - - - - + - - -
Anterior + + + + + + + + + - - - - - - - - - + + - - -
High - - - - - - - - - + + + + + + - + + - - + + + + - - - - + +
Low - - - - - - - - - - - - - - - + - - - - - - - - - + - -
Back - - - - - - - - - - - - + + + + - + - - - + - - + + + + + +
Round - + - - - - + + + +
Nasal - - + - - + - - - + - - - - + -
Continuant - - - - - - - + + - - - - - - - + + + + + +
Voice - + + - + + + + - + - + - + + - + + - + - - + + + + + + + + + +
(Tense) + - + - - - + - + -

Note t ha t in t he above t ab l e , a f fr i c at e s and p l o s iv e s are n o t d i f f e r ­

ent i a t e d in t erms o f a maj or c la s s featur e . In t he Chom s ky -H a l l e s y st em ,
t he s e t wo sound c la s s e s are d i st ingui shed b y r e l e a s e f e at ur e s . A fr i ­
c at e s are [ +D e l ay e d r e l e a s e ] and p l o s ive s are [ -De layed r e l e a se ] . Since
i n B I affricat e s ( l e I and I J / ) are b o t h palat a l and p l o s ive s are a l l
non-p a l a t a l s o und s , t he p l a c e featur e s coronal , anter ior and back are
enough to ke ep t he t wo sound c la s s e s di s t inc t .
 CHAPTER 3

POSITIVE COND I TIONS

The c on c e p t o f ' Po s it ive Cond it i on s ' w i t hi n a generat ive phonology

framework wa s f ir st sugge s t e d by S t a n l e y ( 1 96 7 ) , who introduced t he
idea ( al o ng w i t h t he i d e a s of ' If -Then C ond it i on s ' and ' Negat i ve Con­
di t ions ' ) to e x pr e s s t he c anon i c a l s hap e s o f und e r ly i ng forms o f morph­
eme s . In t he ir d e s c r i p t i o n of t he phono l o gy o f Akan , S c ha c ht er and
Fromkin ( 1 9 6 8 ) s t at e general c on s t raint s on the s y l l a b l e struc t ur e o f
Akan i n t erm s o f ' Po s i t iv e Cond i t i on s ' ( c f . Hyman 1 9 7 5 : 1 1 0 -1 1 2 ; Brown
1 9 7 2 : 5 5 , et c . ) . Brown ( 1 9 7 2 ) i n her d e s c r i p t i o n o f t he p hono logy o f
Luma saaba u s e s ' Po s i t ive Cond it ion s ' t o s t at e sequent i a l c o n st r aint s on
t he sequence of p hono l o gi c a l e l ement s ( c on sonant s and vowe l s ) w i t h i n
p hono l o g i c a l w o r d s and l e x i c a l ent ri e s , and on t he s e quence o f segm e nt s
in t h e p ho ne t i c rea l i sat i o n w i t h i n a p hono l ogic a l t one .
In t he pre sent d e s c r ip t ion , ' Po s it ive Cond it ion s ' are u s e d to c ap t ur e
t he canon i c a l shap e s o f unde r l y i ng form s o r morpheme s ( embodying p o s s i b l e
morpheme s ) i n B 1 . The c anon i c a l shap e s render informa t i o n ab out t he
g eneral c onstraint s on t he s equenc e o f s egment s ( c on sonant and vowe l s )
i n t he p hono l o g i c a l repre sent at i on o f wor d s and l e x i ca l e n t r i e s . To
some degr e e , t he g eneral const raint s a l so h o l d t rue in t he p ho ne t i c
l
real i sa t i o n wit hin p hono l o g i c a l wor ds . B y a ' p hono l ogi c al word ' i s
meant t he max imum unit wit hin whic h a s s im i l at i on and s y l lab i f i c at ion
proc e s s e s m ay o c cur . In many i n st an c e s , phon o l o g i c a l wor ds are c o ­
e x t e n s i v e w i t h grammat i c a l word s , i . e . uni t s or rorms re sult ing from
c ombin ing s t em mor pheme s ( le x i c a l format ive s ) wi t h a f f i x morpheme s
( gr ammat i c a l format i v e s ) . The n o t i o n o f ' phono l o g i c a l wor d ' may b e
somewhat une a s i l y r e l at ed t o t he tradit ional d e f i n it ion o f t he t erm

l
Consonant c lusters may occur in word-initial position on the phonet ic level but not
on t he phonological level .

39
40

' word ' whic h t end s t o r e st o n c r i t er i a su c h a s ' i solab i l i t y ' , ' un i n ­

l
' c o he s ivene s s ' , e t c .
t errup t ab i l it y ' ,
Words ( in t he sense of comb ina t i on s o f morpheme s ) in BI are c on s t r u c ­
t ed b y m e a n s o f a f f ixat ion and/or redup l i cat ion . There ar e however
morphem e s t ha t never t ake any a f f i x e s or undergo redup l ic at io n , such a s
d e n g a n [ d a l) a n ] ' wi t h ' , k e t i ka [ ka t i ka ] ' wh e n ' , di [di] 'at, in, on ' ,
ke [ ka ] ' t o' , d a n [ d a n ] ' and ' , e t c . A s f ar a s t he i r st ruc t ur al pat t ern s
are c onc erne d , t he y are s im i l ar e i t he r t o t ho s e o f s t em morphemes or t o
t ho se o f a f f i x mor p heme s . Thus t he canon i c a l forms o f wor d s ( i . e .
morp hem e c omb inat ion s ) and l e x i c a l ent r i e s c an a l so be c ap t ur e d b y
st at ing t he c anon i c a l shap e s o f s t em a n d a f f i x morp heme s .
This c hapt er con s i st s o f four sec t i on s . The fir st s e c t i o n d ea l s
w i t h p o s i t ive cond i t ions t hat expr e s s perm i t t ed seque n c e s o f p hono l ogic a l
s e gment s w i t hi n l e x i c a l s t em s . The se c o nd s e c t i o n d ea l s wi t h p o s i t i ve
c ond i t i on s t hat e xpr e s s p erm it t ed sequenc e s o f p honolo g i c a l segment s
w i t hin a f f i xe s . The t hird s e c t i on pr e sent s examp l e s o f und e r ly ing forms
of t yp i c al word s , i . e . wor ds c on st ru c t e d by mean s o f a ff ixat ion and/or
r ed up l i c at ion . The four t h s e c t i o n pr e s ent s t he ana l y s i s of t yp ic a l
nou n s whi c h a r e a c c ount e d f o r i n an a d hoc way i n ear l ier s t ud i e s o f B I .
I n s t at ing p ermit t ed sequenc e s o f p hono l o g i c a l segment s w i t h i n mor­
p heme s or word s , t he f o l lowing symb o l s wi l l b e u sed :

+ format ive b oundar y , ++ word bo undary

C [ -S y l l a b ic ] (i.e. c o n sonant s and g l id e s )
V [ +S y l l ab ic ] ( i . e . vowe l s )

3 .1. L E X I CA L STEMS

Hartmann and S t o r k ( 19 7 2 : 2 1 9 ) d e f ine t h e t erm s tem a s " That whic h i s

le ft o f a word when a l l i n f l e x ional - a f f i x e s have b e en remove d .
The st em o f a word i s t he form usua l l y l i st e d in d i c t ionari e s . " A st em
whi c h i s ult imate ly i rredu c i b l e i s c a l l e d a lex ical stem or root ( s ee
Mat t he w s 1 9 7 4 : 4 0 , 7 3 ) . A lex i c a l st em c on s i st s o f one morpheme onl y ;
henc e t he t erm simple s tem i s o f t en used t o re fer t o a l e x i c a l s t em a s
o p p o s e d t o a complex stem ( i . e . a s t em whi c h c on s i st s o f one root and
one or more a f f i x e s ) or a compound stem ( i . e . a stem whic h c o n s i st s o f ,
at l ea st , t wo r o ot s ) .
N o t e t hat int er s e c t ing parent he s e s in t he fo l l o wing formula are used
t o ind i c at e t ha t at l ea s t one o f t he it ems ( el ement s ) b e t ween t he

l
c f . Lyons 1968 : 1 99 , 205 ; also Matthews 197 4 : 20-24 and the references c ited therein;
Brown 1972 : 31 , etc .
 41

parent h e s e s i s ob l igat ory . And i f a morp heme i s gr eater i n lengt h t han

t hat c haract e r i sed i n t he s equent i a l s t at ement , t he c ondit i o n s w i t h i n
brac ket R m u s t r e app l y .
The c anon i c a l s hap e s o f l e x i c a l stems i n B I c a n b e formulat e d a s
f o l l ow s :

++ ( e )( V ( e ( e ) ) V ( e R « e ) v ( e ) ) R ) ++

T h i s formula st at e s t hat :

( a ) a l e x i c a l stem c on s i s t s m i n ima l l y o f ev or VV

( b ) t he max imum sequence o f e i s e e , and t h i s e e s equenc e may o c c ur

only in medial p o s it i o n

( c ) t he max imum s eque n c e o f V i s VV

( d ) t he l e x ic al s t em ( if p o l y s y l lab i c ) may b eg i n w i t h ev , evv , evev,

eveev, vv, vev , o r veev

( e ) a l e x i c a l st em may end e i t her in e or V

The oret i c al l y , t he r e are 4 2 d i f ferent t yp e s o f mor pheme st ructure

t hat c an be der ived from t he above formu l a w i thout reap p l y ing c ondit i o n s
within bracket R . However , t he c or p u s ( c f . S e c t i o n 1 . 3 . ) , b e ing l im i t e d
i n na t ur e , fai l s t o c ontain i t e m s t hat e x emp l i fy a l l t he 4 2 p o s s ib l e
morp heme st ructure s . The unexemp l i f ie d stru c t ur e s are e i t her p er ipheral
(i.e. having a very low frequenc y o f o c c ur re n c e s i n a ct ua l morpheme s )
o r are s imp l y c o n c e ivab l e morp heme s t r u c t ur e s i n t hat t he y c on form t o
t he gener a l c on st ra in t s enumer a t ed above wi t hout any a c t ua l i t ems
e x i s t ing in t he l anguage whi c h exemp l i fy t hem . Thi s is part i c u l a r l y
t rue w i t h p o l y sy l l a b ic s t r uc t ur e s ( i . e . st ruct ur e s whic h c on t a i n t hr e e
o r m o r e Vs e a c h ) w hi c h have t wo c o n s o nant c lu s t e r s . T h e y are ex emp l i ­
f i ed o n l y b y l oan items s u c h a s i n d u s t r i / i nd u s t a r i / [ i n d u s t r i J ' in du s ­
try ' ( Du t c h/Eng l i s h ) , i n t e n s i f / i n t an s i f / [ i n t an s � f . i n tans �pJ ' i n te n ­
s i ve ' ( Du t c h/Eng l i s h ) , s em o a h y a n g / s am b a h y a Q / [ s am b a h y a Q ] ' to say a
p r a y e r ' ( Arab ic ) , et c . who s e s t ruct ur e s are ve eveevev , veeve eve and
eveeveeve re spec t ive l y . Lex i c a l s t em s w i t h t he mono s y l lab i c struct ure s
ev and eve a l so b e long t o t he p e r i p heral type . The st ruc t ur e ev for
inst anc e , is ex emp l i f ie d only b y one l e x i c a l s t em t o my knowledge , v i z .
/yu/ [yuJ ' s ha r k ' . Thi s i t em may take c e rt ain a f f ix e s l i ke / l a h /
' emphat ic suf f i x ' . / k a h / ' int errogat i ve s u f f i x ' . / b a r / ' adj e c t i v i ser
prefix ' , e t c . In addit ion , t he i t em , b e ing a noun , c an be p l ura l i se d
by redup l i c at ing it .
It i s genera l ly agre e d b y s c ho la r s o f Austrone s ia n l i ngui s t i c s t ha t
t he maj o r it y o f l e x i c a l s t e m s i n Austrone s ian l anguag e s ( t hu s , a l so i n
42

l
B I ) are b i sy l l ab i c ( i . e . c ont a in ing t wo V s eac h ) . Thi s c la im i s
suppor t e d b y t he dat a . A samp l e o f 2 0 2 s t ems o f 2 5 6 wor ds ( i . e . uni t s
o r forms c on s i st ing o f l e x i c a l st ems p lu s a f f ix e s and/or redup l i c a t i o n }
found in t he dat a ( c f . S e c t i o n 1 . 3 . ) shows t hat 1 8 9 it ems ( 9 3 . 1 p e r
c ent ) a r e b i s y l l a b i c , and o f t he remain ing 1 3 i t em s , 1 2 i t ems ( 6 . 4 p er
cent ) are t r i sy l l ab i c , and one i t em ( 0 . 5 p e r c ent ) i s mono s y l lab i c .
S t ru c t ural l y , t he 1 8 9 i t ems c an b e groupe d i n t o 1 1 t y p e s and t he 1 2
t r i s y l l ab i c i t ems i nt o five t ype s . The d i st r i b ut ion o f t he 1 8 9 i t ems
over t he 11 t yp e s of s t r u c t ur e shows a c o n s i derab l e rang e o f d i sp er s al .
The s t r u c t ure eveve for e x amp l e i s exemp l i fi e d by 9 0 it ems ( le x i c a l
st em s ) , whi l e t he st ruc t ur e VV i s exemp l i f i e d o n l y b y o n e le x i c a l stem ,
viz . / i a/ C i a] ' he , she ' . ( As a mat t er o f fact , t h i s i s t he only
morpheme i n BI t hat I am aware o f hav ing t he struc t ure VV . ) Thu s i n
s hort , w i t h i n t he b i sy l la b i c group i t s e l f t he r e i s a wide range o f
freque n c y o f o c c urrenc e s or u s e o f st ruc t ur e s i n t he l anguage .
It must b e n o t e d however t hat t he 1 7 t y p e s o f st ructure ( one mono ­
s y l l a b i c , 11 b i s y l l ab ic and five t r i sy l l ab ic ) whi c h appear in t he 2 0 2
s t ems o f t he 2 5 6 samp l e word s ment ioned above d o not ne c e s sari l y imp l y
t hat t he y a r e a l l more c ommon t han t ho s e whi c h happ e n n o t t o b e e x em ­
p l i f ied i n t he samp l e . Nor do e s t he number o f i t ems e x emp l i fying e a c h
o f t ho se struc t ur e s nec e s sari ly c orre spond t o t he propor t i o n o f a c t u a l
morp heme s in B I . The struc t ur e s VV and eve f o r e x amp l e , are exempl i f i e d
by o n e l e x i c a l st em ea c h , v i z . / i a/ C i a ] ' he , s he ' and / z a m / [ z a m , sam ,
J a m ] ( a b ound root appearing in t he form z a m z a m ' ho ly w a t e r ( from
M e c ca ' s s ac r e d we l l ) ' ) . The st ruc t ure eve however ha s a highe r fr e ­
quenc y o f oc currenc e s t ha n t he stru c t ur e VV d oe s . I n t he data ( c f .
S e c t i o n 1 . 3 . ) t he r e a re nine more i t em s o f t he eve s t ruc t ur e whi c h may
t a ke a f f i x e s and/or undergo redup l icat i o n . They are ( 1 ) / s a h / [ s a h ,
� a h.] ' le ga l ' (e.g. / m a n + s a h + k a n / [ m a l) a s a h k a n , m a l) s a h k a n ] ' t o lega l i se ' ) ,
( 2 ) / h a k/ [ h a k , h a ? ] ( e . g . /bar+hak/ [ ba r ha k , barha ? ] ' to ha v e r i g h t ' ) ,
( 3 ) / b a k/ [ b a k ] ' tr o ug h ' (e.g. / b a k+ r a / [ b a k r a ] ' t he trough ' ) , ( 4 ) / ra k/
[rak, ra?] ' s he l f ' ( e . g . / r a k + ra k / [ r a k r a k , ra ? r a ? ] ' s he l v e s ' ) , ( 5) /pak/
[ pa k] ' pa c kag e ' (e. g. / d i +p a k / [ d i pa k ] ' to b e p a c k e d ' ) , (6) / t i k/ [ u k ]
' t o typ e ' ( e . g . / m a n + t i k / [ m a l) a t L k ] ' to do typing ' ) , (7) / t eh/ [ te h ]
' tea ' ( e . g . / t e h +m u / [ t e hm u ] ' your t ea ' ) , ( 8 ) / J am/ [ J a m ] ' w a tc h ' (e.g.
/ b ar+Jam/ [barJam] ' to wear a (wr i s t ) wa t c h ' ) and ( 9 ) / z a i / [ z a t , sat ,
Jat] ' su b s t a n c e ' (e.g. / zat+ra/ [ za tra , sa tra , J a tp a ] ' i t s s ub s ta n c e ' ) .
F inal l y , i t i s ne c e s sary t o p oint out t hat t he c ond i t ions enumerat e d
above , e sp e c i a l l y t he c on dit i o n s on c on sonant s equenc e s hold true mainly

l
See e . g . Halim 1974 : 181ff and the referenc es c ited therein .
 43

wi t h i n nat ive B I it ems . The r e ar e quite a number o f r e c ent l y b o rrowed

i t em s ( le x i c a l st em s ) whic h do not meet t he c o nd i t i o n s enumerat ed above
in t he sense t hat none o f t he ir p honet i c a l t ernant s conform to t h o se
c ond it ion s . Some o f t he l oan it em s c on t a i n c o n s onan,t c lu s t er s o c cur ­
ring in morpheme -init i a l p o s it ion su c h as k l i n i k [ k l i n L k ] ' a Un i a '
( Dut c h/Engl i s h ) , p r a k t e k [ p r a k t e: k ] ' praa t i ae ' ( Engl i sh/Dut c h ) , t r a n s po r
[ t ransp�r] ' transpor t ' ( En g l i sh/Dut c h ) , e tc . , and some c o nt a i n seque n c e s
o f t hree o r more c o n so nant s suc h a s s t r u k t u r [ s t r u k t o r ] ' s t r u a t ur e '
( Engl i s h/Dut c h ) , i n s t r uks i [ i n s t r u ks i ] ' in s trua t i o n ' ( Dut c h/Engl i s h ) ,
i nspektur [ i nspektor ] ' in sp e a tor ' ( Dut c h/Eng l i s h ) , et c . The se i t em s
have t o b e r e p r e s e n t e d i n s u c h a way on t he phono log i c a l l e v e l so t hat
t hey c on form to t he g eneral c ond i t i o n s enumerated above . Inadmi s s i b l e
seque n c e s ( c o nsonant c l u s t er s ) wi l l be broken u p b y i n sert ing a s c hwa .
Thus c l u s t er s in morphem e -init i a l p o s i t ion w i l l be broken up by i n s e r ­
t ing a sc hwa immed iat e l y a f t e r t he f ir st c o n sonant , a n d cons onant
c lu st er s c ont a in ing t hree or four c on sonant s i n morpheme -m e d i a l po s i t i o n
w i l l b e broken up b y insert ing a s c hwa b e t ween t he sec ond a n d t he t hird
c o n sonant s . A c c or d i ng l y , t he l oan items c it e d above w i l l b e p hono­
log i c a l l y r epre sent ed as ( 3 . 1 ) b e l o w :

(3.1) kl i n i k: / k e l i n i k/ [ k l i n i k ] ' a Un i a '

p r a k t e k : / p a r a k t e k / [ p r a k t e: k ] ' p r a a t i ae '
s t r u k t u r : / s a t r u k t u r / [ s t r u k to r ] ' s tr ua tur e '
i n s t ru ks i : / i n sat ruks i / [ i n s t ru k s i ] ' i n s tr u a t i o n '
i n s p e k t u r : / i n s ap e k t u r / [ i n s p e k t o r ] ' i n sp e a t o r '

( See a fur t her d i sc u s s i o n o f t he se l o anwords in S e c t i o n 5 . 6 . )

3.2 . AFFI XES

The c anon i c a l s hap e s o f a f f i x e s i n B I c an b e formulated a s f o l lows :

+ ( C ) V( C ( V » +

The formula st at e s t hat :

( a ) an a f f i x c on s i s t s m inima l l y o f V

( b ) t he max imum s equenc e o f C i s C

( c ) t he max imum s e quen c e o f V i s V

( d ) an a f f i x may end e i t her in C or V

BI make s u s e o f 2 3 ( p honologi c a l ) a f f ix e s for i n f l e c t ional and der i -

vat i onal purpo se s . The funct i o n and meaning o f a n a f f i x dep e nd l ar g e l y
upon t he st em t o whi c h it i s at t ac hed . Take f o r i n s t ance t he pr e f ix
/ ba r -/ . W i t h verb s t em s , t he pr e f i x / b a r - / i s i n f l e c t ional ( i . e . it
44

s imp l y marks t he synt act i c al func t i on o f t he stems wi t hout c hanging

t he gramma t i c a l c la s s or c ategory o f t he s t em s ) denot i ng i n t r an s i t ive
verbs a s i n ( 3 . 2 ) ; with noun st ems , it is der ivat ional ( i . e . i t c hang e s
t he gramma t i c a l c l a s s o f t he stems from n o un int o verb o r adj e c t ive )
who s e meaning may d i f fer from stem to s t em as in e xamp le s ( 3 . 3 ) b e l o w .

( 3 . 2 ) / be r - / + verb stem
/ b e r + p e p i l [ b <:> r p e p i ] ' to s i ng ' ( int r . )
/ b <? r + t e l) k a r / [ b e r t e l) ka r ] ' to qua rre l '
/ b e r + ke l a h i / [ b e r ke l a i ] ' to be invo l v e d in a fig h t '
/ber + dandan/ [ bardandan] ' to ma k e up o n e s e l f '
/ b e r + s em b u p i / [ b e r s em b u p i ] ' to h ide (onese lf) '

The c orre spond ing t r an s it ive form s o f t ho se ver b s are , respec t iv e l y :

/men+p a p i + k a n / [mepap i ka n ] ' to s i ng a song ' , /men+per+ten k a r+ k a n
[ m e m p e r t a l) k a r k a n ] ' to quarre l ( up on ) ' , /men+per+ke l a h i +kan/
[mampe r k e l a i kan] ' to fi g h t (on s om e thing ) ' , /men+d a n d a n + i / [ me n da n da n i ]
' to b e a u ti fy ' , / m e n + s em b u p i + k a n / [ m a p a m b u p i k a n ] ' to hi de ( s ome th i ng ) ' .

( 3 . 3) /ber-/ + noun stem

/ b e r + s a p e d a / [ be r s ep e d a ] ' to go by B I CYCL E '
/ b e r + f o t o / [ b a r fo t o ] ' to ha v e one ' s PICTURE ta k e n '
/ b a r + l um p u r / [ b a r l umpo r ] ' to be MUDDY '
/ber + a i r/ [ bara L r ] ' to be WATER Y '
/ b a r + ka t a / [ ba r k a t a ] ' to say WORDS '

A d e t a i l e d a c c ount o f t he fun c t i o n and mean ing o f t he s e 2 3 a f f i x e s

woul d ent a i l t he d i s c u s s i on o f synt ax and t hi s i s c er t a i n l y b e y ond t h e
sc ope o f t hi s s t udy . S u f f i c e it t o p o int out here that i n B I , who se
vo c a b u lary is re lat i ve l y smal l er compar e d wi t h modern European language s ,
a f fixat i o n i s a v ery s ig n i f i c ant p henomenon in word c on st ruct i o n . As
a r e s ult , most ( le x i c a l ) s t ems c an t ake a number o f affixe s . Examp l e s
( 3 . 4 ) b e l o w may su c c inct l y i l l u s t r at e t h i s p henomenon . A l l of t he
word s given b e low have t he same root ( l ex i c a l s t e m ) v i z . /bua t/ ' do ,
make ' .

( 3 . 4 )1 . / be r + BUAT/ [ b a r b u a t ] ' to ac t'

/ d i + B UAT/ [ d i b ua t ] ' to be made/do n e by '
/ m a n + B U A T / [ m am b u a t ] ' to m a k e , to do '
/ t a r + B UA T / [ t a r b u a t ] ' to be made of, mana g e d to do '
/ BUAT + k a n / [ b u a t k a n ] ' m a k e 80m e thing for s o m e b o dy '
/ BUAT + l a h/ [ b ua t l a h ] ' do ( i t) p l e a 8 e '
/ BUAT + a n i [ b u a t a n ] 'produc t '
 45

2 . I p a + m a n + B U A T I [ p am b ua t ] 'maker '
I p a + m a n + B U AT + a n i [ p a m b u a t a n ] ' t he ma k i n g . pr o du c t io n '
I p a r + BUAT + a n + p a l [ p a r b u a ta np a ] ' hi s de e ds '

Table 2 b e low shows t he t yp e s o f struc t ur e and t he t yp e s o f a f fi x e s

( pr e f i x e s , s u f f i x e s and i n f i x e s ) in B I .

T ABLE 2

Structure Prefix S u f f ix I n f ix Tot a l

l. ev d i , ka , ku , k U , mu , pa 8
pa , sa
2. eve b a r , m an , k a h , 'k a n , l a h , 9
par , ta r p u n , wa n
3. evev wa t i 1
i
4. V i 1
5. ve an a I , am , a r 4

Total 9 11 3 23

I t mu s t b e p o inted o ut ho we ver t hat some o f t h e forms ( af f i xe s )

l i st ed i n Tab l e 2 above repre s e nt t wo or mor e morphemes ea c h . The form
I l a h l for ex amp le can b e an ' emp hat i c s u ff i x ' as i n I k a m u + l a h m a k a n l
[ kamu l a h ma k a n ] ' Yo u who s ho u Z d e a t ' , or a ' mi l d impera t ive forming
su f fi x ' a s in I k a m u m a k a n + l a h l [ k a m u m a k a n l a h ] ' ( Yo u ) ea t p Z e a s e ' , e t c .
It mu st a l s o be p o in t ed out , furthermor e , t hat t he f o rms g i ven in
Tab l e 2 above are p hono l o g i c a l ( und erl y ing) form s o f affixe s ( morpheme s )
and t here for e prec l ude a l lomorp h s .
A s t o t he operat ive func t ion o f t he s e a f f i x e s , one can s t a t e b r i e f l y
t ha t t he y di ffer i n t he i r pr oduc t iv it y . Some a f f i x e s s u c h as I m a n l
' ac t ive v o i c e v erb p r e f i x ' , I d i l ' pa s sive vo i c e verb pre fix ' , Ibarl
' i ntran s it ive/st at ive verb forming pr e f i x ' , I k a n l ' t ran s i t i ve verb
form ing s u f f i x ' , I i i ' t ran s i t ive verb forming s u f f i x ' , 1 e t c . are c apab l e
o f c omb i ning w i t h a large numb er o f i t em s , and some can c omb ine only
wit h a handful o f i t em s . The c a s e s in p o int are Iw a n l ' ma l e human noun
forming suf f i x ' , I w a t i l ' female human noun forming su f f i x ' and t he
infixes - I - a I - I , I - am - I and I - a r - I . Not e t hat t he s e i n f i x e s are no
l onger u s ed produc t i v e l y in c ont empor ary B I . The oc currence of t he s e
i n f i x e s i n mo dern B I i s l imit e d t o a number o f long- e st ab l i shed i t e m s
or p hra s e s suc h a s t e l u n j u k [ t a l u P J o ? ] ' fo r e fi ng e r ' ( + I - a l l + I t up J u k l
[ tuPJO? ] ' t o p o i n t ' ) , g em u r u h [ g am u r u h ] ' t hunde r i ng ' ( + I - am i + I g u r u h l

l
The suffix Ikanl i s different synt actically from the suffix I i i in that I kanl may
have an indirect object while I i i can only have a direct object .
46

[gu ruh] ' t h under ' ) , t e r a l i [ ta r a l i ] ' l a t t i ce ' ( + I - a r - I + I t a l i l [ t a l i ]

'rope ' ) , k em u n i n g [ k a m u n i Q ] ' y e l low ' ( + I - am - I + I k u n i Q I [ k u n i Q ]
' y e l l ow ' ) , k e l e l aw a r [ k a l a l a wa r ] ' ba t ' ( + I - a l - I + I k a l a w a r l [ k a l a w a r ]
' ba t ' ) , et c . A l t h ough- t he s e i n f i x e s can st i l l b e ident i f i e d i n a num­
b e r of l ong-e s t ab l i s hed wor d s and p hr a s e s , t he y show no c lear patt ern
of mean ing and func t i o n . I t is therefore j ust i fi ab le to t r e at t he forms
c o n s i st ing of l e x i c a l s t em s and i n f i x e s c it e d abo ve and the l ike a s
' l ex i c a l i sed ' items . ( See a l so Tab l e s 4a-c for furt her informa t ion o n
t he produc t iv i t y o f t he se affix e s . )

3 . 3. E X AM P L E S O F U N D E R L Y I N G F O R M S OF T Y P I CA L WO R D S

I t ha s b e e n stat e d above for t he purpo se o f t he pr e s ent de sc ript ion

t hat a ' p hon o l o gi c a l word ' _ i s t he max imum unit wit h i n whic h a s s im i l at i o n
a n d s y l l a b i fi c at ion pr o c e s s e s may o c c u r . A p ho n o l o gi c a l word may furt her
be c harac t er i s ed b y st r e s s p l a c ement whi c h , general l y , oc curs on t he
penul t imate s y l l a b l e o f t he word . Thi s sect ion , whi l e exemp l i fy ing a
number of ( und e r l y i ng ) s t r u c t ur e s o f l e x ic a l stems c aptured by the
formula s e t up in S e c t ion 3 . 1 . , p r e s e nt s e x amp l e s o f und e r l y i ng forms
of t yp i c a l p ho no l o g i c a l word s , t hat is wor d s whi c h st r u c t ural l y c o n s i st
o f s t ems p lu s affi x e s and/or d up l i c at i on o f st ems .
It ha s b e e n po int ed out in Sect i on 3 . 1 . t hat in t he data t her e are
2 5 6 p hono l og i c a l wor d s c on s t ruc t e d out o f 2 0 2 l e x i c a l s t ems . I t ha s
al so b e e n s t at e d t hat t he 2 0 2 l e x i c a l s t ems exemp l i fy 1 7 d i f fe rent t yp e s
o f s t ru c t ur e . Let u s e xamine t he d i s t ri but i o n o f t he 2 0 2 l e x i c a l st ems
over t he 17 t y p e s of s t r u c t ure . Lex i c a l st ems in t he f o l lowing e x amp l e s
ar e c ap i t a l i sed .

a) 90 st ems are o f t he struc t ur e CVCVC ( [ -Sy l l , +Syl l , -Sy l l , +Sy l l ,

- S y l l ] ) a s in ( 3 . 5 ) :

(3. 5)1. I B O D O H + i l [ bo d o h i . b o d o i ] ' t o outwi t, to foo l someone '

2. IT I T E L + fl a l [ t i t e l fl a ] ' hi s /her/i t s t i t l e '
3. I ka + B E BAS + a n I [ ka b e b a s a n . kab aba s a n ] ' fr e e dom '
4. I m a n + R AM A Y + k a n l [ m a r a m a y ka n ] ' t o ce l e b r a t e '
5. I p a + m a n + D I D I K + a n I [ p an d i d i k a n ] ' e du c a t io n '

b ) 3 9 stems are of t he st ruc t ure CVCV ( [ -S yl l , +Syl l , - S yl l , +Syll ] )

a s i n ( 3 . 6 ) b el ow :

(3. 6)1. IDURI + anI [dur ian] ' a t horny fru i t '
2. I m a n + G U LA + i l [ m a Q g u l a i ] ' t o sugar '
3. I m a n + T <3 M U + k a n l [ m a n am u k a n ] ' to di s co v er '
4. Ipa + DESA + anI [ padesa?an . padasa ?an] ' v i l la g e area '
5. I p a + man + TAR I I [ pana r i ] ' da n ce r '
 47

c ) 1 7 st ems are of t he structure CVCCVC ( [ -Syl l , +Sy l l , -Sy l l , -Syl l ,

+Syl l , -Syl l ] ) as i n ( 3 . 7 ) b e low :

(3. 7)1. Isa + BaR S I HI [ s abars i h ] ' a s c lean ( a s ) '

2. I tar + PANOAYI [ ta r panday] ' t h e c l e v er e s t '
3. I m a n + P I N O A H + k a n l [ m am i n d a h k a n ] ' to remove '
4. I ka + BANJ I R + ani [ kabapJ i ra n ] ' to be f l o o de d '
5 . I d i + p a r + T U..N J U K + k a n l [ d i p a r t u p } u k a n ] ' to b e s hown '

d) 1 3 st ems are o f t he s t ru c t ure CVVC ( [ -Sy l l , +Sy l l , +Syl l , -Sy l l ] )

a s i n ( 3 . 8 ) b e l ow :

( 3 . 8 ) 1 . Isa + LUASI [ s a l uas] ' a s b i g as '

2 . I B UAT + a n i [ b u a t a n ] 'pro duc t, trade ma r k '
3. I k a + BA I K + a n i [ k a b a i k a n ] ' ki n dne s s '
4. I m a n + p a r + B A l K + i l [ ma m p a r b a i k i ] ' to repa i r '
5 . I p a + m a n + M UA Y + a n i [ p am u a y a n ] ' expan s i on '

e ) 8 st ems are of t he s t ru c t ur e VC CVC ( [ +Sy l l , - Sy l l , -Syl l , +Sy l l ,

- S yl l ] ) as in ( 3 . 9 ) be low :

( 3.9)1. I A � G U K + a n i [ a l) g u k a n ] 'no d '

2. I b a r + .3M PAT I [ b a r am pa t ] ' i n ( b y ) four '
3. I d i + AM B I L + p a l [ d i a m b i l p a ] ' ta k e n b y h i m/ her '
4. I ka + I N OAH + a n i [ ka i nd a ha n ] 'beauty '
5 . I U N O A � + u n d a l) l [ u n d a l) u n d a l) ] ' law, regu l a t i o n '

f ) 7 st ems are of t he s t ruc ture VCVC ( [ +Sy l l , -Syl l , + S y l l , -Syl l ] ) a s

in ( 3 . 1 0 ) b e l ow :

( 3 . 10 ) 1 . I m a n + A J A R I [ m a l) a } a r J ' to t e ac h '
2 . I A O I K+m u l [ a d i km u ] ' y our ( y o unge r ) b r o t her/s i s t e r '
3. I t a r + I �AT I [ t a r i l) a t ] ' t o remember ( i n c i de n t a l ly ) '
4. I s a+ O R A I:j I [ s ao r a fl ] ' a p e r so n '
5 . IAX I R+P aI [ ax i r p a . a h i r p a . a k i r p a ] , t he e nd, a t l a s t '

g) 6 s t ems ar e of t he struc t ure CVC CV ( [ -S y l l , +Sy l l , -Sy l l , - Sy l l ,

+Sy l l ] ) a s i n ( 3 . 1 1 ) b e l ow :

( 3 . 11 ) 1 . I m a n + G A � G U I [ m a l) g a l) g u ] ' to di s t urb '

2. I b a r + H a NT I I [ b a r h a n t i ] ' to s t o p '
3. Id i +T .3 R KA I [ d i t a r ka ] ' t o be g ue s s e d '
4. I ba r + Ka R J A I [ b a k a r } a ] ' to wor k '
5 . I d i +sa RTA+ i l [ d i s a r t a i ] ' t o b e a ccompan i e d '
48

h ) 5 st ems are o f t he struc t ure vev ( [ +Sy1 1 , - S y1 1 , +Sy1 1 ] ) :

( 3 . 12 ) 1 . / m e n + I S I / [ m e l) i s i ] ' to fi H '
2. / A D A+ l a h / [ a d a l a h ] ' t ha t / i t i s '
3. / I BU+pa/ [ i bupa] ' hi s/heY' m a t he Y"
4. / A P A+ k a h / [ a p a k a h ] ' is it, what '
5. / A KU + p u n / [ a k u po n ] ' I too '

i ) 5 s t em s are o f t he struc t ure evvev ( [ -Sy 1 1 , + S y l l , +Sy l l , - Sy l l ,

+Syl l ] ) .

( 3 . 13 ) 1 . / S UAM I +p a / [ s u a m i pa J ' heY' husban d '

2. / S UARA+pa/ [ s ua r a r a ] ' heY'/hi s v o i c e '
3. / s e+ S U A T U / [ s e s u a t u ] ' some t hing '
4. / S I A P A+ k a h / [ s i a p a k a h ] 'who '
5. / B I A SA+p a / [ b i a s a p a ] ' u s ua H y '

j ) 3 st ems are o f t he s t ru c t ure eveevev ( [ - S y l l , +Sy l 1 , -Sy 1 l , -Sy l l ,

+Syl l , -Syl 1 , +Syl 1 ] ) - s e e ( 3 . 14 : 1 ) , 2 st ems are o f t he s t r u c t ur e
veevev ( [ +Syl l , -Sy l l , -Syl 1 , +Syl l , - S y 1 l , +Syl 1 ] ) - s e e ( 3 . 14 : 2 ) ,
2 st ems are o f t he struc t ure evevev ( [ - S y l l , +Sy l 1 , - S yl l , +Syl l , -Sy l l ,
+Sy l l ] ) - s e e ( 3 . 1 4 : 3 ) , 1 stem i s of t he st ruc t ure VV ( [ + S y l 1 , +Sy l l ] )
- s e e ( 3 . 1 4 : 4 ) , 1 s t em i s o f t he s t ruc t ur e vve ( [ +Syl l , +Sy l 1 , -Sy l l ] )
- s e e ( 3 . 1 4 : 5 ) , 1 st em i s o f t he struc t ure veev ( [ + S yl l , -Sy l 1 , -Syl l ,
+Sy l l ] ) - s e e ( 3 . 1 4 : 6 ) , 1 s t em i s o f t he struc t ure eveveve ( [ - Sy l l ,
+Sy l l , - S yl 1 , +Syl 1 , -Syl l , +Sy 1 l , - S yl 1 ] ) - s e e ( 3 . 1 4 : 7 ) , and 1 stem
i s of t he struct ure eve ( [ - Sy 1 l , +Sy l 1 , -Sy1 l ] ) - see ( 3 . 14 : 8 ) .

( 3 . 1 4 ) la . / S d ND I R I + an/ [send i r i an ] ' a lone '

b. / ke + M <3 R D E KA + a n / [ k e m e r d e k a ? a n ] ' independenc e '
c. / T d R T AWA+ k a n / [ t e r t a w a k a n ] ' to l a u g h a t '
2a. / I S T <3 R I + p a / [ i s t e r i p a ] ' hi s w i fe '
b. /ANTARA+p a/ [ a n t a r a p a ] ' among a t heY' s '
3a. / b e r + K <3 L A H I / [ b e r k e l a i , b e r k e l a h i ] ' to fi g h t '
b. / B <3 RA PA+ k a h / [ b e r a p a ka h ] ' how many/muc h '
4. / I A+ l a h / [ i a l a h ] ' he / s h e /i t i s '
5. /A I R+pa/ [ a i rpa] ' t he wa t e Y"
6. / b e r +A R T I / [ b e r a r t i ] ' to m e a n , meani ng fu l '
7. / N E G AT I F + r a / [ n e g a t i f p a , n eg a t i p p a ] ' t he ( fi lm ) n e g a t iv e '
8. / Z A M + z a m / [ z a m z a m , s am s a m , J a m J a m ] ' ho ly wa t e Y"

T h e struc t u r e s o f t he 2 0 2 l ex i c a l st ems pre s ent ed above c an b e

summar i se d a s fo l l ow s :
 49

TABLE 3

Number of Re ference to
St r ucture Percentage
i tems Examples

Mono syl labic :

1. eve 1 0.5 ( 3 . 14 : 8 )
Bi syl labic :
2. eveve 90 44 (3.5)
3. evev 39 19 . 3 ( 3 . 6)
4. eve eve 17 8.4 (3.7)
5. evve 13 6.4 ( 3. 8)
6 . ve eve 9 4.4 ( 3 . 9)
7 . veve 7 3.5 ( 3 . 10 )
8. eveev 6 3 ( 3 . 11 )
9 . vev 5 2.5 ( 3 . 12 )
1 0 . vv 1 0.5 ( 3 . 14 : 4 )
II. vve 1 0.5 ( 3 . 14 : 5 )
1 2 . veev 1 0.5 ( 3 . 14 : 6 )
Trisyl labic :
13. evvev 5 2. 5 ( 3 . 13 )
14 . eveevev 3 1.5 ( 3 . 14 : 1 )
1 5 . veevev 2 1 ( 3 . 14 : 2 )
1 6 . evevev 2 1 ( 3 . 14 : 3 )
17 . eveveve 1 0.5 ( 3 . 14 : 7 )

2 02 100

The c e nt ral p o i nt i n t h i s s e c t i o n s o far ha s b ee n t o e x emp l i fy t he

und er l ying structures o f t he 2 0 2 l e x i c a l s t em s whic h , by virtue o f
a f fi xat i on and/or redup l i c at io n , make u p 2 5 6 phono l o g i c a l wor d s . The
i n c r e a se i n t he number of words c ompare d wit h t he number o f l e x i c a l
st ems i nd i cat e s t hat c e r t a in l e x i c a l stems t ake d i f ferent a f f i xe s and/or
undergo r edupl i c at i on .
The f o l l o wing t ab l e s ( 4 a , 4 b , 4 c ) show t he o c c urren c e o f a f f i x e s and/or
redup l ic a t i o n i n t he 2 5 6 phono l o g i c a l word s . Not e t hat t he numbering
of a f f i x e s of t he same t ype o c c urring i n suc c e s s ion is from l e ft to
r ight . T hu s t he pr e f i x e s I m an - I and I p a r - I , and t h e suf f i x e s I - k a n l
and I - p a l i n t he word m em p e r t i m b a n g k a n n y a I m a n +p a r + t i m b a Q + k a n + p a l
[ mampa r t i m b a Q k a n p a ] ' to con s i der i t ' for examp l e , are i dent i fi e d a s
Prefix 1 , Prefix 2 , Suffix 1 and Suffix 2 r e sp e c t ive l y .
Tab l e 4 a b e l ow shows t he oc currenc e o f pre f i x e s and redup l i c at io n
( RE D ) in 1 1 4 words whi c h do n o t t ake a n y s u f f i x e s .
50

TABLE 4a

Pre fix 2

Prefix 1 0 man par RED Total

0 - - -
8 8
ba r 13 - -
5 18
di 7 - 1 -
8
ka 1 - - -
1
man 35 - 4 6 45
pa 2 5 - -
7
par 2 - -
- 2
sa 15 - -
- 15
tar 10 -
- - 10

Total 85 5 4 19 114

T a b l e 4 b b e low shows t he o c c urrenc e o f s u f f i xe s in 1 2 4 words . Eac h

word has a suffix w i t h or wit hout a pr e f i x or redup l ic at i o n ( RED ) .

TABLE 4b

Suffix
Prefix , Total
RED an i ka h kan ku l ah mu pa pun

0 15 3 2 3 1 9 1 35 2 71
bar 2 - -
- -
2 - -
- 4
di -
1 - 2 -
- -
1 -
4
ka 25 -
- - - - - - -
25
man - 3 -
9 - - - -
- 12
pa 1 - - -
- - - - -
1
par -
2 - -
- 2 - -
- 4
RED 3 - - - -
- - - -
3

Tota l 46 9 2 14 1 13 1 36 2 124

Tab l e 4 c b elow shows t he oc curr e n c e o f a f f i x e s and redup l i c at ion

( RED ) i n 1 8 wor d s - t he r emaining samp l e wor d s , whi c h c annot b e c ap t ur e d
i n t ab l e s 4 a and 4 b above . Dot s a r e u s e d t o ind i c at e t ha.t the p r e f i x
o n t h e l e f t appl i e s t oget her w i t h t he it ems ( a ffixe s or redup l i c at io n )
o n t h e t o p o f t he c o l umns o f t he number s . The numb er s t he m s e l v e s show
t he frequency of t he oc currenc e s o f t he a f f i x e s on t he top o f t he c o l umns
t oget her w i t h t he pr e f i x e s o n t he l e f t .
 51

TABLE 4c

Pre fix 2 Suf fix (1 , 2 )

P r e f ix 1 man par RED an kan pa Total

ba r . . . . . . . . . . . .1. . . . .1 1
di · . . . . . 4 • • . . . . . • • . . . • • . . . . 4 4
ka · . . . . . . . . . . . . . . . . . 1. . . . . . . . . . .1 1
man · . . . . . 3. . . . . . . . . . . . . . 3 3
• • • • • • 4 • • • • • • • • • • • • • • • • • • 4 4
. . . . . . . . . . . . . . . . . . . . . . . . .1 . . . .1 1
pa .4 . • . . . • . . . . . . . . . . 4 4

Tota l : 18

Tab l e s 4b and 4c above show t hat pr e f i x e s may c o - o c c ur with most

su f f i x e s in B I . I n fac t , some pr e f i x e s always func t io n t o get he r s im ­
u l taneou s l y w i t h c er t a i n s u f f i x e s i n t he c on s truc t ion o f c ert ain t yp e s
o f word s . T h i s kind o f p a i r s o f p r e f i x and s u f f i x m a y b e r e f e rred t o
as ' s imu lt aneou s a f f i x e s ' ( s im u 1 f i x ) . T h e mo st produ c t ive p a i r s are
I k a . . . a n I ' ab st r ac t noun forming s imu1 f i x ' , I p a . . . a n I ' noun forming
s im u 1 f i x ' , I p a r . . . a n I ' noun forming s imu1 f ix ' and I p a r . . . kanl
' c ausat ive trans i t ive verb fo rm i ng s imu1 f i x ' . The s im u 1 f i x I k a . . . anI
may b e added t o adj e c t ive s t em s (c f . 3 . 5 : 2, 3.8:3, 3.9:4, 3 . 1 4 : 1 b ) and
t o c ert a in noun s t em s ( c f . 3 . 7 : 4 ) . The s imu l f i x I p a . . . a n I may b e
added t o c e r t a in ver b st ems ( c f . 3 . 5 : 5 , 3 . 8 : 5 ) and t o c er t a in noun
1
st ems ( c f . 3 . 6 : 4 ) . The s imu 1 f i x I p a r . . . a n I may be added t o c er t a i n
st em s s u c h a s p e r s e ko l a h a n [ p a r s a k o l a h a n ] ' sc ho o l s y s t e m ' ( + I s a ko l a h l
[ sa ko l a h ]
ma t t e r s '
' sc ho o l ' +
(+ l u s a ha l [ usaha]
Ipar . . . anI ) , p e r u s a haan [ p a r u s a h a ? a n ]
' e nt erp r i s e , effo r t ' + I p a r
' e n t erpr i s e
. . . a n I ) , etc .
and t o c er t a i n v er b stems su c h a s p e r t em u a n [ p a r t a m u a n ] ' m e e t i ng '
( + I t am ul [ t am u ] 'me e t ' + I p a r . . . a n I ) , pe r t e n g ka r a n [ pa r t a8 k a r a n ] 'a
quarre l ' ( + I t a 8 k a r l [ t a 8 k a r ] ' to q uarre l ' + I p a r . . . a n I , etc . The
s imu 1 f i x I p a r . . . k a n l may be added t o c e rt a in noun stems s u c h a s
p e r soa l ka n [ pa r s o a l ka n ] ' t o rai s e q u e s t i o n on ' ( + I soa l l [ soa l ] 'ques­
tion ' + I p a r . . . ka n / ) , p e r ha t i k a n [ pa r h a t i ka n ]
( + I h a t i l [ ha t i ] ' hear t ' + I p a r . . . k a n / ) , e t c . , and t o c e r t a i n verb
' t o p a y a t t e n t i o n to '

s t em s s u c h a s p e r t u n j u k k a n [ m a m p a r t u p l u k a n ] ' to s h o w ' (+ I t u p l u k l
[ t uplo?] ' to s how, p o in t ' + I p a r . . . k a n/ ) , pe r d e n g a r k a n [ p a r d a8 a r ka n ]

l
Not e however that the examples ( 3 . 5 : 5 , 3 . 8 : 5 ) both have Imanl in the underlying rep­
re sentat ion.
52

' t o p l ay (mus i c ) fo r o t he r s t o l i s te n to ' ( + / d a C) a r / [ d a C) a r ] ' to l i s t e n '

+ / par . . . k a n / , etc .
T a b l e 4 c s hows furt hermore , that c er t a i n wor d s may ha ve t hr e e a f fi x e s
(i.e. t wo p r e f i x e s and one suffix or one p r e f i x and t wo suf f i x e s ) e ac h .
Ther e are i n fac t a number o f word s t ha t have more t han three a f f i x e s
eac h s u c h a s p e n d i d i k a n n y a l a h [ p a n d i d i k a n p a l a h ] ' i t i s h i s/ h e r educa ­
t i on ' ( + / p a +m a n + D I D I K+ a n + p a + l a h / ) , d i p e r t u n j u k ka n n y a l a h
[ d i pa r t up J u ka n p a l a h ] ' t h e n i t was s ho wn by him/he r ' ( + / d i + p a r + T U N J U K+
ka n+pa+ l a h / ) , etc .

3 . 4. A N OTE ON WOR D A N A L Y S I S

It w i l l have been not i c ed in t he ex ampl e s given in Sect ion 3 . 3 . above

t ha t t he pr e f i x /m a n / p o s t ulated in t he und er l ying repre sent a t i o n o f a
c er t ai n t yp e o f wo rd s - word s c ont a in ing t he pr e fi x / p a / w i t h or w i t ho ut
t he su ffix / a n / , s u c h as p e n a r i ' dancer ' ( c f . 3. 6 : 5) , p e n d i d i ka n ' e du ­
ca t i o n ' (cf. 3. 5:5) , p e m ua i a n ' expa n s i on, e x t e n s ion ' (cf. 3 . 8 : 5 ) , etc .
i s p hone t i c a l l y r e a l i s ed e it her as a nasal or ' z ero ' dep e nd ing upon the
sound t ha t imm ed iat e l y f o l lows it . The n a s a l r e a l i sat ion o f t he / m a n /
pr e f i x o c c ur s when i t i s immed iat e l y fol l owed b y a n o b s truent or vowe l ,
and t he ' z ero ' r ea l i s a t i o n oc c ur s whe n it i s immediat e l y fo l lowed by
t he o ther t yp e s of so und s - nasal s , l i quids and semi -vowe l s .
Work s on B I t hat I am aware o f s o far make no p o s t u l at i o n o f t he
pref ix / m a n / in t he anal y s i s o f word s o f t he type c it ed above . The
common prac t i c e among s c h o l a r s wor king on B I is to analyse t he t hr e e
word s above a s p e N + t a r i , peN + d i d i k + an and peN + m u a i + a n r e sp e c ­
t iv e l y , t hen t o d e sc ribe how t he morpheme ( pr e fi x ) p e N - ha s several
a l l omorphs whic h are p hono l og i c a l l y d e finab l e . The n a s a l segment ( N )
in t he pre f ix p e N - i s homorganic w i t h s t op s , fric at ives or vowe l s
f o l l o wing i t , and i t b e c ome s ' z ero ' b e fore the other t yp e s o f sound s .
I t wi l l s o on b e c ome c l ear t hat t hi s s ort o f anal y s i s i s not p hono lo gi c ­
a l l y p l a u s ib l e . It fa i l s t o a c c ount for t he pa ir s o f word s in ( 3 . 1 5 )
b e l o w wher e t he pr e f ix p e - in a - i t ems and t he pr e f ix p e N - i n b -i t em s
bot h have t he same func t ion , i . e . t o f o rm an agent noun from a v e r b s t em .
N o t e t hat t he a - it em s in t he fo l l owing e x amp l e s u s ua l l y treated a s
e x c e p t ions t o t he na s a l a s s im i lat ion proc e s s d e s c r i b e d above . ( The
ana l y s i s given b e l ow is b a s e d on t he surf a c e s t ru c t ur e o f t he word s . )

( 3 . 1 5 ) la . p e k e r j a [ pa k a r J a ] ' a wor ker ' ( + p e + k e r j a ' t o wor k ' )

b. p e n g e r j a [ p a l) a r J a ] ' t he e x e c u t o r of a j o b/p l an ' ( + p e N + k e r j a
' t o wor k ' )
2a. pe s e r t a [ pa s a r t a ] ' a par t i c i p an t ' ( + p e + s e r t a ' to par t i c i p a t e ,
' t o accompany ' )
 53

b. penye r t a [ papa r t a ] ' t he company ' (+ peN+s e r t a ' t o p a r t i c ipa t e ,

to acc ompany ' )
3a. pes u r u h [ pa s u r u h ] ' a jani tor ' ( + p e + s u r u h ' t o orde r , t o s e n d
s om e b ody to ' )
b. p e n y u r u h [ p ap u r u h ] ' t h e p e r s o n who ordered/s e n t s o m e o n e '
( + peN+s u r u h ' to order/send bo dy (to) ' )
4a. pet i nj u [ pat i pJ u ] ' a b o xe r ' ( + p e + t i n j u ' t o fi s t ' )
b. pen i nj u [ pan i pJ u ] ' t he p e r s on w h o p u n c h e d ( s omeone ) '
( + p e N + t i n j u ' to fi s t ' )
5a . p e d a g a n g [ p a d a g a l) ] ' a trader/b u s i n e s sman ' (+ pe+d a g a n g ' to
trade ' )
b. p e n d a g a n g [ p a n d a g a l) ] ' t he p e r s on who s o l d goods ' (+
peN+dagang ' t o t rade ' )

To re gard p e - and p e N - i n t he above e xamp l e s a s two d i f fe rent p r e ­

f i x e s ( morpheme s ) wi l l b e a ga i n st t he i n t u i t i ve knowledge o f t he nat i ve
speakers about t h e i r language , e s p e c i a l l y t he knowledge o f words t hat
t he generat i ve phono l o g i s t s e e k s to d e s c r ib e . Mo st nat i ve speakers o f
B I wi l l fe e l t hat t h e di fference i n me aning i n e a c h o f t he above p a i r s
o f w o r d s ( nouns ) i s not muc h subj e c t t o the p r e f i x e s p e - a n d p e N - but
rat her t o the type o f ve rb s t hat e a c h o f t ho s e p a i r s i s a s s oc i at ed w i t h .
The a-it ems ( i . e . items w i t h the p r e f i x p e - ) in e x amp l e s ( 3 . 1 5 ) above
are usua l l y a s so c i at e d w i t h verbs c on s i s t ing o f verb st ems p lu s / b a r - /
' st at ive/adj e c t ival verb p r e fi x ' (e.g. 3 . 1 5 : 1a , 2 a , 4 a , 5 a ) o r verb st ems
p l u s /d i - / ' pa s s ive vo i c e verb pre f i x ' (e . g . 3 . 1 5 : 3a ) . The b - items
(i.e. i t ems w i t h the p r e f i x p e N - ) on t he other hand , are usual l y a s s o c i ­
at ed w i t h ve rb s c on s i s t ing o f verb s t ems p lu s /m a n - / ' ac t i ve v o i c e
( t ran s i t ive ) verb p r e f i x ' w i t h or wit hout t he pre f i x / - k a n / or / i / ( b o t h
a r e tran s i t i ve v e r b forming suffixe s ) . Thu s when a nat ive s p e aker o f
BI i s asked t o d i f fe rent i a t e for e xamp l e t he p a i r p e k e r j a or p e n g e r j a
(cf. 3 . 1 5 : 1a , lb ) , he i s l ik e l y t o de fine t h o s e items by saying , more or
l e s s , t hat p e k e r j a ' wo r k e r ' has somet hing t o do with (a p e r s on who )
b e ke r j a ' ( t o ) work ' (+ /ba r/ + / k a r J a / ' to work ' ) and p e n g e r j a ' t h e
e x e cu tor of a j o b /p l an ' h a s some thing t o do w i t h ( a person who )
me n g e r j a k a n ' ( t o ) exe c u t e /do ( s ome t h i n g ) ' ( + / m a n / + / k a r J a / ' t o work/
do ( root ) ' . In a s imi l ar way , he wi l l d i fferent i a t e p e s u r u h ' a jan i tor '
(cf. 3 . 1 5 : 3a ) from p e n y u r u h ' t h e p e r s o n who orde r e d or s e n t someone '
(cf. 3 . 1 5 : 3b ) by saying t hat p e s u r u h has some t hing t o do w i t h ( a person
who i s ) d i s u r u h ' be orde re d/s e n t ' ( + /d i / + / s u r u h / ' to order/send
(to) ' ) , and p e n y u r u h has some t hing t o do w i t h (a p e r s o n who ) m e n y u r u h
' ( to ) order/s e n d someone ' ( + /m a n / + / s u r u h / ' to order/send someone ' ) .
54

The problem re l at e d t o t he ( surface ) pre f i x e s p e - and p e N - i n the

above e x amp l e s ( 3 . 1 5 ) i n whi c h b o t h p e r form t he same fun c t ion , c an b e
so lved b y e s t a b l i shing two di fferent s t ru c t ur a l patt erns in t he unde r­
l ying re p r e s e n t at ion . The t wo t ype s of i t ems i l lu s t rated by e x amp l e s
( 3 . 1 5 ) ab ove c an be repre sent ed , re spe c t ive l y , b y t h e phrase markers
( s t ru c t ure s ) ( a ) and ( b ) in ( 3 . 1 6 ) b e l ow .

( 3 . 16 ) a . NP b. NP

�P �vp
I
P pr

pe-
I STEM
I pr� v

pe-
I
meN-
I
STEM

( lpa/) ( l p a / ) ( lm a n / )

N o t e howe ver t hat t h e s t ruc t ure ( a ) above repre sent s items ( nouns ) whi c h
are usua l l y a s s o c i a t e d w i t h ve rb s c ontaining / b a r - / o r /d i - / , but s in c e
t ho s e pre f i x e s n e v e r a p p e a r on the s u r f a c e forms one finds no part i c ular
need t o p o s t u l a t e t hem i n the unde r l ying repre sent at i on . On t he o t he r
hand , t he s t ruc t ure ( b ) repre sent s it ems ( nouns ) whi c h a r e u s ua l l y
a s s o c i a t e d w i t h ve r b s containing t he p r e f i x /m a n / , and alt hough t he
p r e f i x does not always appear on the s urface forms , t he ' z ero ' re a l i sa­
t i on o f the pre f i x , a s we saw early in this s e c t ion , is phono logi c a l ly
de finab le .
In a s imi lar way one c an a c c ount for t he p a i r s o f i t ems in ( 3 . 1 7 )
b e low whi c h are c lo s e l y related t o t he p a i r s given ( 3 . 1 5 ) above . The
a - i t e m s in the fo l l owing e xamp l e s are c o n s t ru c t e d by adding / p a r . . . a ni
' noun forming s imul f i x ' t o ( verb ) stems , and the b - i t ems are c on s t r u c ­
t e d b y adding / p a . . . a n i ' noun forming s imul fix ' t o verb s b e ginning
with ( p o s t ulated ) /m a n / . Like i n e xamp l e s ( 3 . 1 5 ) , the a-it ems in t he
fol l owing examp l e s are a s s o c i a t e d w i t h verbs be ginning w i t h the p r e f i x
/ b a r / , a n d the b - i t ems a r e a s s o c i a t e d w i t h ve rb s b e gi nning w i t h t he
p r e f i x /m a n / .

( 3 . 1 7 ) la . p e k e r j a a n [ p a k a r J a ? a n J ' w o r k , occupa ti o n ' ( + / k a r J a / ' to

work ' ( root ) + /par . . . ani)
b. p e n g e r j a a n [ p a l) a r J a ? a n J ' t h e execu t i o n o f a j o b /p l an '
( + ( /m a n / ) + / k a r J a / ' t o work ' + /pa . . . ani)
2a . pe s e r t a a n [ pa s a r t a ? a n J ' p a r t i cipa t i o n ' ( + / s a r t a / ' to par t i ­
c i p a t e /acc ompany ' ( root ) + /pa r . . . a n i )
b. pen y e r t a a n [ papa r t a ? a n J ' t h e a c t o f acc o mpany i n g ' ( + ( /m a n / )
+ / s a r t a / ' t o accompany ' + /pa . . . a n i )
 55

3a . *persu ruhan
b. p e n y u r u h a n [ p a fl u r u h a n ] ' t h e a c t o f orderi ng/s ending someone '
( + ( /m a n / ) + / s u r u h / ' to order/send ' / p a . . . a n i )
4a . p e r t i n j u a n [ p a r t i fl J u a n ] ' ma t t e r s re l a t e d t o boxi n g ' / t i fl J U /
' t o fi s t ' ( root ) + / p a r . . . ani)
b. p e n i n j u a n [ p a n i fl J u a n ] ' t he a c t o f p u n c h i n g ' ( + ( lm a n / ) / t i fl J U /
' t o fi s t ' + / p a . . . ani)

,
Sa . perdagangan [ pardaga�a n ] ' ma t t e r s re l a t e d t o trade s '
( + /d a g a � / ' t o t rade ' ( root ) + / p a r . . . a n i )
b. pendagangan [ pandagana n ] ' t h e a c t o f trading/s e l l i n g goods '
( + ( /m a n / ) + / d a g a � / ' t o trade ' + / p a . . ani)

Not � c e t hat the pre f i x / p a r / i n e xamp l e s ( 3 . 1 7 : 1a , 2 a ) above i s phone t ­

i c a l l y rea l i sed a s C p a ] b e c au s e i t oc curs b e fore a s y l l ab le c on t a i n i n g
- e r ( c f . S e c t ion 5 . 1 ) . Not i c e furt hermore , t hat t he form * p e r s u r y h a n
i s not a c c e p t ab l e . Nor i s the form * b e r s u r u h . The fact that t here i s
n o verb * b e r s u r u h i n B I w i t h whi c h t he form * p e r s u r u h a n c an b e a s s o c i ­
a t e d d o e s n o t ne c e s sar i l y mean t hat t h e p r e f i x / b a r / needs t o b e p o s t ­
u l a t e d i n t h e underlying repre sentat i on o f words o f t he a-t ype in the
ab ove e xamp l e s . The que st ion whe ther a c e r t a in s t em c an have t he s im­
ulfix / p a r . . . a n i c annot be so lved by the p o s t ulat ion of / b a r / in t he
underlying re pre s e n t at i on o f t ho s e i t ems b e c au s e it has yet t o be de c i de d
whe t her t he stem i n que st i on c an b e pre fixed w i t h / b a r / or not . In
other words , t he a f f i xat i o n of stems w i t h / p a r . . . a n i or / b a r / depends
large l y upon t he type of stems .
The t wo t yp e s o f words i l l u s t r a t e d by e xamp l e s ( 3 . 1 7 ) above c an b e
repre s e n t e d on t he phono l o g i c a l l e v e l b y t he s t ruc t ure s ( a ) and ( b )

!1Pr
l
r e s p e c t i ve l y .

( 3 . 18 ) a . NP b. NP
�
pr nx Pre f i x VP Suffix

pe r - STEM -an pe-

I A I pr

me N -
EM

STEM -an
(lpar/ ) (lan/) ( l p a / ) ( lm a n / ) ( la n / )

Thi s chapter , e s pe c ia l l y the l a s t s e c t i on , has b e e n devoted t o t he

d i s c u s s i o n o f the morphology o f BI . It s r e l evan c e t o t he p honology o f
B I however i s undoubt e d l y s i gn i f i c ant . Moreover , s in c e t he greater part

l
Note however that Ipa . . . ani can be added to a number of noun stems where /man/ i s
not postulated i n the underlying representation ( c f . 3 . 6 : 4 ) .
56

o f a phon o l o g i c a l d e s c ript i on i s c on c erned w i t h words , i t i s d e s i r a b l e

t o deal w i t h morpho l o g i c a l pro c e s s e s t hat a r e re l e vant t o the c l a r i t y
o f the phon o l o g i c a l d e s c ript ion o f t he l anguage in que s t ion .
In t he c urrent generat ive grammar framework , t he d i s c u s s ion o f
morpho l o g i c a l pro c e s s e s ( i f any ) wi l l appear p i e c emea l i n the synt a c t i c
and phon o l o g i c a l c ompone nt s o f t he grammar . The framework de s c ri b e d
by Chomsky ( 1 9 6 5 ) , whi c h h a s b e e n w i d e l y a c c e p t e d by l ingui s t s , appar­
ent l y gives no room for t he d i s c u s s ion o f d e r i vat i onal morpho l ogy in
the grammar . It i s c laimed t hat derive d nomina l s are entered d i re c t l y
int o t h e unde r l y ing s t ru c t ure from t he l e x i con l i ke other format i ve s o r
morpheme s w i t hout d e s c r i b ing t h e p ro c e s s e s o f t h e i r format ion . ( In the
1 9 5 7 ver s i on o f t he generat ive t heory , this t ype of noun i s derived b y
t ran s forma t i o n ru l e s - t h u s t he proc e s s e s o f deri vat ion a r e a c c ount e d
for . ) Whi le the generat ive l ingu i s t s e e k s t o d e s c r i b e the comp e t e n c e
or the t a c i t knowle d ge o f words a n d t he ru l e s t hat govern t he manner i n
whi c h morpheme s a r e p u t t oge t he� t o y i e l d t h o s e word s , the c urrent gen­
e rat i ve framework lacks me c hani sms for t he t re atment of the morpho l o g i c a l
aspect s .
The inappropriate t reatment o f morpho l o gy in the c urrent gene rat ive
framework has l e d some l ingu i s t s ( e . g . Hal le 1 9 7 3 ) t o propo s e a ' word
format ion ' c omponent wi t hi n a generat ive framework , where morphol ogy
c an b e dealt wit h in i t s own r ight . The d i s c u s s ion o f st ems , a f f i xe s
and s ome word c on s t ruc t i on s in t h i s c hapt e r wi l l probab l y give a part i a l
answer t o t h e propo sed ' word format i on ' c omponent .
 CHAPTER 4

MORPHEME STRUCTURE COND I TIONS

A l t hough the n o t i o n o f ' re dundan c y ' has l ong b e en r e c o gn i s e d i n

phono l o g i c a l ana l y s i s ( for e xamp l e i n an inventory o f phoneme s o f a
language where ' re dundant feature s ' are in a way h e l d re spon s i b l e for
the e x i s t e n c e o f variant s or a l l ophone s ) , t he c on c e p t of ' phono logi c a l
redundanc y ' has gained c on s i derab l e s i gn i f i c an c e i n t he t heore t i c a l
framework o f generat ive phono l ogy . It has deve loped a l ong w i t h t he
c o n c e pt o f s imp l i c i t y - a c onc ept whi c h forms part o f gene rat i ve t heory .
Thu s Chomsky ( 1 9 6 2 : 2 3 3 ) wri t e s

. . . we mu s t appar e n t l y do w h a t a n y s c i e nt i s t d o e s w h e n f a c e d
w i t h t h e t a s k o f c o n s t ru c t i n g a t h e o ry t o a c c ount f o r p a r ­
t i c u l a r s ub j e c t mat t e r - - n a me l y , t r y v a r i o u s way s a n d c h o o s e
t h e s i mp l e s t t h a t c an b e f o u n d .

Simp l i c i t y has b e e n made a b a s i s or c r i t e rion for j udging t he re lat ive

me r i t s o f t wo or more comp e t ing de s c ri p t i on s of a part i c ular l anguage .
The e s s ent i a l c laim in gene rat i ve grammar i s t hat t he s imp l i c i t y me t r i c
( al s o c a l l e d ' e valuat i o n p r o c e dure ' i n t h e l i t e rat ure ) mus t b e p art o f
o r b u i l t i n t o t h e theory . Chomsky ( 1 9 6 2 : 2 4 5 ) goe s on t o say

An i m p o r t a n t part of t h i s t h e o r y w i l l b e a p r o c e du r e o f
e v a l u at i o n t h a t w i l l p e r m i t c h o i c e b e t w e e n a l t e r na t i v e
p r o p o s e d g r ammar s f o r p ar t i c u l a r l an g u a g e s .

In pract i c e , t he simp l i c i t y met r i c has b e e n a s s o c i a t e d , i f not

l
e quat e d , w i t h the numbe r of feature s required to c ap t ure a gene r a l i sa­
t ion . A de s c ri p t i on w i t h fewer feature s i s s a i d t o be s imp l e r , more

l
Many generative phonologists however have quest ioned the validity of feature counting
as the only basis for the evaluation of descriptions because it fails to capture the
degree of naturalne ss of sound classes so that , very often , what seems s impler ( i . e .
a class o f sounds that requires fewer features t o spec ify) i s less natural , and what
seems less simple i s phonologically more natural ( c f . Chen 1973 : 226 ; Hyman 1975 : 139-
142 ) .

57
58

natural ( t hus more genera l ) and more favoured t han t he one t hat ut i l i s e s
more feat ure s . It i s t here fore de s i rab l e , Halle ( 1 9 6 2 : 3 4 0 ) rec ommended ,
t hat " . . . we mus t omit feat ure s in a l l d i c t ionary repre sentat i on s , when­
ever t he s e c an b e introduced by a rule t hat is l e s s c o s t l y t han t he
saving it e f fe c t s " . The s e omi s s i b le feat ure s are redundant in t h e s e n s e
t hat t hey c an b e pred i c t e d from o t her ( phono l o g i c a l a n d grammat i c a l )
feature s . Rul e s s t at ing t he s e redundan c i e s are var i o u s l y c a l l e d
' morpheme s t ru c t ure r u l e s ( MSRs ) ' ( Ha l l e 1 9 5 9 ) , ' morpheme s t ru c t ure
c ondi t i on s ( MS C s ) ' ( St a n l e y 1 9 6 7 ) , ' le x i c a l redundanc y rule s ' ( Chomsky
and Ha l l e 1 9 6 8 ) , e t c . The e x i s t e n c e o f the various t e rmino logie s not
only shows t hat wri t e r s on generat ive phonology have devot ed c on s ider­
able at t en t i o n t o the redundan c y aspect in phono l o gy , but also i nd i c a t e s
t hat gene rat ive phono l o g i st s have n o t reac hed any agreement o n t he t ype
of framework for deal ing w i t h t he phono l o g i c a l redundan c i e s . The
approac h adopt ed in the pr esent d e s c ri p t ion is t hat pro p o s e d by S t a n l e y
( 1967 ) .
A ft e r d i s c u s s ing a number o f problems inherent i n MSRs approac h ,
s u c h a s the problem in choos ing one feat ure over t he o t her , t he s t a t u s
o f b lank feat ure , e t c . , Stanley ( 1 9 6 7 ) propo sed t hat M S R s b e r e p l a c e d
w i t h morpheme structure cond i t ions ( MS C s ) . He p o i n t e d o u t t hat MSRs
are b a s i c a l ly s t at i c ' in t hat they do not convert one level o f repre sen­
t a t i on i n t o anot her , but rat her s imp l y s t a t e redundan c i e s on the phono­
l o g i c a l l e ve l . Qui t e to the c ont rary , phon o l o g i c a l rul e s proper ( P _
rul e s ) c onvert phon o l o g i c a l repre s e n t a t i o n s i n t o phone t i c one s . Thu s ,
S t an l e y drew a c ru c i a l d i st i n c t i on b e t we e n a c on s t raint on a g i ven l e ve l
o f repre sentat i on and a rul e t hat c onvert s one l e ve l o f repre sentat ion
i n t o another . Onc e we t ake s e r i o u s l y the not i on t hat a grammar c o n t a i n s
rule s a s w e l l as repre sentat ions , t here i s no r e a s o n t o require t hat
phono l o g i c a l repre s e n t a t i o n s s hould c ontain spe c i fi c at i ons only for a l l
unpr e d i c t a b l e ( non-re dundant ) feature s , so long a s t h e grammar as a
who l e a l l ows us t o det ermine what feat ure s are commutab l e and hen c e un­
pre d i c t a b l e ( non -redundant ) and what f e at ure s are dependent and henc e
pre d i c t ab l e ( redundant ) .
In such t e rms , Stanley propo sed t hat under l ying ( phono l o gi c a l ) re p ­
r e s entat ions , l ike phone t i c repre sentat i on s , s houl d be ful l y spe c i fied
for all feat ure s . An add i t i onal s e t of morpheme s t ru c t ure cond i t i o n s
o r c on s t raint s t hen provide s , f o r a n y g i ven language , a s pe c i f i c a t i o n
o f t he range o f p o s s i b le s e gment s and morpheme s , inc luding dependenc i e s
that may e x i s t feat ure value s . The morpheme s t ruc t ure c ondi t ions ( MS C s )
c ap t ure t wo kinds o f redundanc i e s - sequent i a l and s e gme n t a l redundan­
cies .
 59

The formal requirement s for MSCs are :

( i ) they may not c hange feat ure value s

( i i ) they are unordered
( i i i ) t he y operate w i t h i n , and not a c ro s s , morpheme boundar i e s

In t h e pre sent d e s c ript i on M S C s are numb ered for c onve n i e n c e o f r e f ­

erence a n d d i s c u s s ion o f s im i l ar aspe c t s . The numb e r s do not imp l y any
ordering in app l i c at i on . The y are unordered b e c au s e t h e y may not c h ange
feat ure value s .
The MSCs are e xpre s s e d in t erms o f ' I f-Then c on d i t i on s ' . Eac h p a i r
o f ' I f-Then condit ions ' s t at e s t hat i f a part i c ular c ondi t ion o b t a i n s
i n a c er t a i n envi ronment , t hen some o t h e r c ondit i o n m u s t a l s o be met .
Thi s c hap t e r i s organ i s e d into three s e c t ions . The f i r s t s e c t ion
dea l s w i t h s e quent i a l c ondit i ons wit hin BI morpheme s ; t he s e c ond s e c t i on
d e a l s w i t h s e gme n t a l c ondit i o n s , and t he t hird s e c t ion pre sent s t he
summary o f c ondit i on s formu l a t e d in the first t wo s e c t i on s .

4.1. S E Q UE N T I A L ' I F - T H E N CO N D I T I O N S '

Sequent i a l ' I f-Then Condit ions ' s t a t e predi c t ab l e ( redundant ) feat ure s
re s u l t ing from the c on s t raint s on t he seque n c e s o f phono l o g i c a l seg­
ment s w i t h i n morpheme s . The condi t i o n s formu l a t e d b e l ow are b a s e d
m a i n l y on the c on s t ra int s or c ondi t i ons t hat hold w i t h nat i ve B I ( Malay )
i t ems . However , some regularit i e s in t he s e quenc e s o f phono logi c a l
s e gment s o b s e rved i n b o rrowed i t ems a r e a l so t aken i n t o a c c ount t o the
e x t ent t hat the i t ems have been natura l i s e d in t he s e n se t hat t he y no
l onger sound fore i gn and t hat t hey conform t o t he general c o n s t ra int s
( p o s i t ive c on d i t ions ) s t a t e d in Se c t i on 3 . 1 .
Greek l e t t e r s a , B, y i n t he fo l lowing c on d i t i on s ( and a l s o i n P-ru l e s
in Chap t e r 5 ) st and for t h e p l u s and minus val u e s o f feat ure s , and t hey
indicate agreement o f val u e s i n d i f fe rent part s of a c ondi t i o n ( or rule ) .

4 . 1. 1 . S E QU E NCE O F C O NS O NANTS

It has been p o i n t e d out i n S e c t i on 3 . 1 . that BI a l l ows o n l y t wo c on ­

s onan t s t o o c c ur i n a seque n c e w i t hin nat ive i t ems , a n d s u c h s equenc e s
c an o n l y o c c ur i n word-me d i a l p o s i t ion . B I imp o s e s further re s t r i c t i o n s
on the t ype s o f c o n sonant s t hat may o c c ur i n a s e quenc e . To find out
t he re s t r i c t ions t hat B I s y s t emat i c a l l y e x hi b i t s on t he seque n c e s o f
s e gment s i n t he unde r l y i ng form o f morpheme s , l e t u s f i r s t e x amine c on ­
sonant seque n c e s found i n the dat a ( c f . S e c t i on 1 . 3 .) then forma l i se the
regulari t i e s o b s e rved on t ho s e se quenc e s .
6.0

Of the first t wo hundred morpheme s c ontaining c ons onant s e quenc e s

found i n t h e dat a , 1 4 2 i t ems ( 7 1 per cent ) have sequenc e s o f n a s a l s
p l u s ob s t ruent s ( c f . 4.1) , 2 5 i t ems ( 1 2 . 5 p e r c e n t ) have s e quen c e s o f
t he t r i l l / r / p l us o t her c on s onant s ( c f . 4 . 2 ) and 3 3 it ems ( 1 6 . 5 p e r
c ent ) c ontain var ious kinds o f c on s o nant s e quenc e s (cf. 4.3). A l t hough
the samp l e is sma l l in numb e r , it suffi c ient l y re f le c t s t h e c on s t raint s
t hat BI s y s t emat i c a l l y imp o s e s on c on s onant s e quenc e s .
The fol l owing e xamp l e s ( 4 . 1 ) show t he summary o f t he s e quen c e s o f
nas a l s and o b s t ruent s i n 1 4 2 i t ems in t he samp l e .

(4.1) Sequence Percentage Examples

la o m-p 10 . 5 / am p a t / [ a m p a t ] , foul' '
/p i mp i n/ [ p i m p L n ] ' t o l e ad '
b. m-b 11 /amb i 1 / [ amb L l ] ' t o t a ke '
/gamba r/ [ gamba r ] ' p i c ture '
2a . n-t 16 . 5 /un t u k/ [ u n to ? ] , for '
/g a n t i / [ g a n t i ] , to change '
b. n-d 11 / i n d a h/ [ i n d a h ] ' be a u t i fu l '
/ p e n d e k / [ p e n d E: ? ] , s ho r t '
3a . r-c 2 / l arcar/ [ l arca r ]
v
' fl u e n t '
/kurc i / [ kurc i ] ' ke y '
b. r-J 3 /J a r J i / [ J a r J i ] 'promise '
/ b a r J i r/ [ ba r J L r ] ' fl ood '
4a. I) - k 7 / a l) k a w/ [ a l) k a w ] 'you '
/ m u l) k i n / [ m U l) k L n ] 'po s s i b l e '
b. I) - g 8 / a l) g u k/ [ a l) g o ? ] ' t o nod '
I t i I) g i / [ t i I) g i ] ' ta l l '
5a. n-s 1.5 / i n s a f/ [ i n s a f , i nsap] ' re a l i s e '
/ l a n s u l) / [ l a n s o l) ] , dire c t ( l y ) ,
b. I) - S 0.5 / b a l) s a / [ b a l) s a ] ' na t i o n '

The above e x amp l e s show that when a n a s a l and an o b s t ruent o c c ur in a

s e quenc e , t h e t wo c o n s onant s are genera l l y homorgan i c (cf. 4 . 1 : 1 - 5a ) .
Thi s general rule however doe s not app l y t o i t ems s u c h as / b a l) s a / [ b a l) s a ]
'nation ' (cf. 4 . 1 : 5b ) , / b a l) s a l / [ b a l) s a l ] ' ha l l ' , / t a l) s i / [ t a l) s i ] ' ba r ­
rack ' , / p i l) s a n / [ p i l) s a n ] ' fa i n t e d ' , / b a f) s a t / [ b a l) s a t ] ' ra s ca l ' , e t c .
In o t her word s , t h e seque n c e na s a l p lu s / s / c annot be pred i c t e d b e c au s e
it c an b e e i t he r / - n s - / ( c f . 4 . 1 : 5a ) or / - I) S - / in phono l ogi c a l l y n o n ­
de finab l e environment s . Exc luding t he s e quen c e n a s a l p lu s / s / , t he
regular i t y ob served in ( 4 . 1 ) above c an be formulated as fo l lows :
 61

(1) If f Cons
li Nas a lJ
[ ]
cons
-Son

�1] � ]
-Cont

}
Then Ant Ant
S C or S Cor
Back Bac k

The c o nd i t ion s t at e s t hat i f t he first c o n s onant o f a sequence o f

t w o c o n sonant s i s s p e c i fi e d as a n a s a l ( [ + C o n s onant , +Na s a l ] ) a n d t he
se cond as a non-sonorant , non- c ont inuant c on s o nant ( [ +Cons onant ,
-Sonorant , -Cont i nuant ] ) - t hu s e x c luding n a s a l , t r i l l , lateral and
fr i c at ive c on s onant s - then the t wo c on s o nant s w i l l agree on the feature s
anter ior , coronal and back . ( No t e that t h e c ondit i o n or rule s imp l y
s t at e s we l l - formedne s s o f morpheme s c on t a i n i n g s e quen c e s o f n a s a l s and
non-sonorant , non-c on t i nuant c o n s onant s ; it doe s not imp l y a b l ank­
f i l l ing not i on . )
The f o l l owing e x amp l e s ( 4 . 2 ) s how t h e o c c urre n c e o f t he sequence
t r i l l p lu s ot her c on s onant s i n 25 i t ems i n t h e samp l e .

(4 . 2 ) Sequence Percentage Examples

1. r-b 2 / k a r b a w/ [ k a r b a w ] ' buffa l. o '
/ k o r b a n / [ ko r b a n ] ' victim '
2. r-d 1 /ma r d e k a / [ m a r d e k a ] ' i ndependen t '
/ma r d u / [ m a r d u ] 'nice ( o f v o i ce ) '
3· r-g 1 /ha rga/ [ ha r g a ] 'price '
/pa rg i / [ pa r g i ] ' to go '
4. r-J 0.5 /ka rJa/ [ ka r j a ] ' to w o r k '
5. r-m 0.5 /pa rma t a / [ pa rma t a ] ' j e we l. '
6. r-n 2.5 /wa r n a / [ w a r n a ] ' c o l. o u r '
/pu r nama/ [ p u r n ama ] ' fu l. l. moon '
7. r- 1 0.5 /p a r 1 u / [ p a r 1 u ] ' to need '
8. r-t 2.5 /a r t i / [ a r t i ] 'meaning '
/sar ta/ [ s a r t a ] ' t o ac company '
9. r-k 1 / t a r ka/ [ t a r ka ] ' t o gu e s s '
/pa r ka r a / [ pa r ka r a ] 'matters '
10 . r-s 1 /b a r s i h/ [ b a r s i h ] ' c l. e an '
/ku r s i / [ ku r s i ] ' c ha i r '
11 . r-c 0.5 /pa r c a ya/ [ pa r c a y a ] ' t o be l. i e v e '

The ab ove e x amp l e s show t hat / r / may p r e c e d e almo s t any t ype of c on s o n ­

ant s . Thi s c an e a s i l y b e unde r s t o o d b e c aus e , as we s a w i n S e c t ion 3 . 2 . ,
62

B I h a s t hree pre f i x e s ending i n / r / ( i . e . / b a r / , / p a r / and / t a r / ) whi c h

c an b e added t o s t ems be ginning w i t h any c ons onant s o r vowe l s .
A furt her e xaminat ion o f BI wo rds shows t hat o f t he v o i c e d c on s o n ­
ant s , o n l y n a s a l s , tri l l and , t o some degree , l a t e r a l may o c cur as t he
first e l ement of c on sonant sequenc e s irre s p e c t ive o f whet her the s e c ond
e l ement is a vo i c e d or voi c e l e s s c onsonant . The o c c urre n c e o f t he
lateral as t he f i r s t e l ement o f c on s onant s e quenc e s i s limited t o l oan­
word s , such as / k a l b u / [ k a l b u ] ' he a r t ' ( Arab i c ) , /a l p a / [ a l p a ] 'absent '
( Dut c h ) , /a l t a r / [ a l t a r ] ' a t tar ' ( Du t c h/Eng l i s h ) , /a l k o h o l / [ a l k o h � l ]
' a t c o ho t ' ( Dut c h/Eng l i s h ) , e t c . Other vo i c e d c onsonant s may a l s o o c c ur
as t he first e l ement o f c on sonant sequenc e s but the s e c ond e lement mus t
a l s o b e a voi c e d c onsonant ( se e t he di s c u s s i on o f e x amp l e s ( 4 . 3 ) b e l ow ) .
The c on s t raint s di s c u s s e d ab ove c an be forma l i s e d as f o l l ows :

(2) If f Cons f Cons

G- Vo i c eJ J
l.:-Vo i c e
Then
!
[ +Son ]

The c on d i t i on ( ru l e ) s t a t e s t hat i f t he s e c ond s e gment o f a sequence

of t wo c o n sonant s i s a vo i c e l e s s c on sonant ( [ +Consonant , -Voi c e ] ) and
t he first s e gment is a vo i c e d c o nsonant , t hen it mu st be a sonorant
s ound ( [ +Cons onant , +Voi c e , +Sonorant ] ) - i . e . a n a s a l or t r i l l or lat ­
era l .
Examp l e s ( 4 . 3 ) b e l ow show t he summary o f t he various kinds o f c on ­
sonant s e quenc e s o c curring in 3 3 it ems in t he samp l e .

(4.3) Sequence Percentage Examples

1. 5 -t 3 /i star i / [ i s tar i ] ' w i fe '
/pa s t i / [ pa s t i ] ' de fi n i t e '
5 -I 0.5 /a s l i / [ a s l i ] ' g en u i n e '
2. k-t 2 /wa k t u / [ w a k t u ] ' t ime '
/d o k t e r / [ d o k t E r ] ' do c tor '
k-s 1 /pa k s a / [ pa k s a ] ' t o force '
/ I aksana/ [ I aksana] ' tike '
k-d 0.5 /d a k d i k / [ d a k d L k ] ' fr i g h t e n e d '
k-n 1 / I a k n a t/ [ I a ? n a t ] ' p un i s hme n t '
/ma k n a / [ ma ? n a ] ' i mp t i c a t i o n '
k-I 0.5 / t a k l u k/ [ t a ? l o ? ] ' s urre n de r '
k-r 0.5 /a k r a p/ [ a k r a p ] ' i n t i ma t e '
k-z 0.5 / t a k z i m/ [ t a ? z L m , t a ? s L m ] 'respe c t '
3. p-t 1 /s a p tu/ [ s a p t u ] ' Sa t urday '
/ s a p t em b e r / [ s a p t e m b E r ] ' Se p t embe r '
 63

Sequence Percentage Examples

4. h-t 0.5 / s aJ a h t a r a/ [ s aJ a h t a r a ] ' p r o s p e ro u s '
h-k 0.5 / b a h k a n/ [ ba h k a n ] 'even ' ( c onj )
h-� 0.5 /d a h � a t / [ d a h s a t , d a h � a t ] ' e normou s '
h-b 0.5 / s a h b a n d a r/ [ s a h ba n d a r ] 'port
o ffi c ia l. '
h-l 0.5 /a h 1 i / [ a h I i ] ' exp e r t '
h-y 0.5 / s a m b a h y a l) / [ s a m b a h y a l) ] ' pray '
h-w 0.5 /ba hwa/ [ b a hwa ] ' that ' ( c onj )
5. s-h 0.5 /ma s h u r / [ m a s h e r ] ' famo u s '
6. m-r 0.5 /zam r u t/ [ za m re t , J a m re t ] ' emera t d '
m- l 0.5 /J u m l a h / [ J u m l a h ] ' to ta l. , amoun t '
7. l -m 0.5 / i l mu/ [ i l mu ] ' science, know t edge '
8. b-r 0.5 /feb r ua r i / [ fa b r ua r i , pabrua r i ]
, Feb ruary '

The f i r s t five groups o f t h e above c on sonant s e quenc e s ( 4 . 3 : 1 -5 ) s how

that when t h e first e l ement ( c ons onant ) of a sequence of two c on s onant s
i s vo i c e le s s , then the s e c ond e l ement can be e i t he r a vo i c e d or a vo i c e ­
l e s s c on sonant . The y show no part i c u lar regularit y wh i c h c a n be i n t r o ­
duc e d by a rul e . The l a s t t hree groups ( 4 . 3 : 6 -8 ) on t he ot her hand ,
s how t hat t h e first and t he s e c ond e lement s o f t ho se c on s onant s equenc e s
agree o n t he feat ure voice . They a l s o sugge s t t hat t h e s e c ond e l ement
o f a seque n c e o f t wo consonant s mu st be a voiced sonorant c on sonant when
the f i r s t e l ement is a vo i c e d c on sonant . This is t rue w i t h c on s onant
s e que nc e s who s e first e lement is a non-sonorant v o i c e d c on s onant a s in
( 4 . 4 ) b e low .

( 4 . 4 ) /fabrua r i / [ fa b r ua r i . pab r ua r i ] ' Fe bruary ' (cf. 4 . 3 :8)

/kod r a t / [ kod r a t ]
/ag ra r i a/ [ a g r a r i a ]
/h i J r ah/ [ h i J r a h ]
/ t a b l e t/ [ t a b l E t ]
/s i gn a l / [ s i gna l ]
' na ture , de s t i n y '
' agri c u t ture '
' t o e v a c ua t e '
' tab t e t '
' s i gna l. '

Not e however that the p a l a t a l and velar n a s a l s / p / and / 1) / respe c t ive l y

never f o l l ow imme d i a t e l y v o i c e d c onsonant s . Note furt hermore , t hat
mo s t , if not a l l , of t he i t ems appearing in ( 4 . 3 ) and ( 4 . 4 ) ab ove are
l oanword s .
The c on s t raint on t he sequence o f t wo vo i c e d c on s onant s d i s c u s se d
ab ove can be formu l a t e d as f o l lows :
64

[
J
(3) If Cons [ +Con s ]

[1 ]
+Vo i c e
- Son

Then VO i C e
+Son
-High

The condi t i on s t at e s t hat if the first e lement of a s e quen c e of t wo

c on s onant s i s s pe c i fi e d as a non-sonorant vo i c e d c on sonant ( [ + C o n s onant ,
+Vo ic e , -Sonorant ] ) , t hen t he s e c ond must be a non-high sonorant c on ­
sonant ( [ +Consonant , +Vo ic e , +Sonorant , -High ] ) . The feat ure s pe c i fi ­
c a t ion [ -H i gh ] b l o c k s the nasal s I p l and I � / .
In ( 4 . 3 ) ab ove , we find no c on s onant seque n c e who s e first e l ement i s
lei . Thi s i s not be c au s e o f t he limitedne s s o f the s amp l e . The sequence
lei p l u s other c on s o nant s doe s not o c c ur on the phono l o g i c a l leve l o f
repre s ent at ion in B I . Cond i t ion ( 3 ) above on t he other hand s t at e s t hat
the s equen c e IJ/ plus non-back sonorant c on s onan t s is admi s s i b l e . The
f o l l owing c ondit ion pre d i c t s the o c c urren c e o f I J I and b l o c k s , in a way ,
t he o c c urre n c e of lei as the first e lement o f c on s o nant s equenc e s .

(4) If +Cons [ +Cons ]

-Son
-Cont
-Ant
-Cor
-Back
�
Then [ +Vo i c e ] f± Vo i c el
li s on J
The c ondit ion s t a t e s t hat i f t he fi r s t e lement of a s e quenc e o f t wo
c on s onant s i s an a ffri c at e ( [ +Consonant , -Sonorant , -Cont inuant ,
-Ant e r io r , -Corona l , -Bac k ] ) t hen it mus t be voi c e d ( /J / ) , and t he
s e c ond w i l l be a v o i c e d sonorant c o n sonant ( [ +Consonant , +Vo i c e ,
+Sonorant ] ) .
Examp l e s ( 4 . 3 : 4 ) above show t hat o f the t hree p hono l o g i c a l g l i d e s
p o s t u l a t e d in S e c t ion 2 . 1 . o n l y I h l o c c ur s a s the f i r s t e lement o f c on ­
sonant s e quenc e s . Again , t h i s n o t b e c au s e o f the l imitedn e s s o f t h e
s amp l e . The b i labial and p a l a t a l g l i d e s ( /wl and / y / respe c t ive l y )
never o c c ur a s the first e l ement o f c on s onant s equenc e s i n B I . Thi s
c on s t raint c a n b e forma l i s e d as fo l lows :
 65

(5) If i- Con sl i- s y l ll

[ ]
l.:- S y l lJ l.:- ConsJ
� �
Then VO i C e [ +Vo i c e ]
+Low
+Cont

The c o nd i t i o n s t at e s t hat i f t he f i r st s e gment o f a s e quenc e o f t wo

non- s y l lab i c s e gment s i s s pe c i fi e d as a g l i de t hen it mu st be t he vo i c e ­
le s s g l o t t a l fri c at ive / h / ( [ -Cons onant , - S y l l ab i c , -Vo ic e , +Low ,
+Cont i nuant ] ) . And i f the s e c ond s e gment is furt her spe c i fi e d as a
g l i de then it mus t be v o i c e d ( i . e . /w/ or / y / ) . The s e c ond part o f t h e
c o nd i t i on b l o c k s t h e o c c urrence o f h - h whi c h i s i nadmi s s i b l e . ( In
fac t , B I does not al l ow sequenc e s of t w o i dent i c a l c o n sonan t s o r g l i de s . )

4. 1.2. S E QU E NCES O F VOWE L S

I n S e c t io n 2 . 2 . s i x vowe l s were p o s i t e d for t h e repr e s e n t a t i on o f

morpheme s o n t h e p hono l o g i c a l leve l . And i n S e c t i o n 3 . 1 . w e saw t hat
B I a l l ow s only two vowe l s at t he maximum to o c c u r in a sequence w i t h i n
morpheme s . The oret i c a l l y , t here are 3 6 p o s s i b l e vowe l s e quenc e s . A
furt he r e xaminat ion o f B1 morpheme s howeve r , s hows t hat some o f t he
p o s s i b l e s e quenc e s ( e . g . s equenc e s c on t a i n ing t he c entral mid vowe l / a / )
do not o c c ur at a l l , and some ( e . g . s e quenc e s c on t a i n ing a front or
back mid vowe l , sequenc e s o f two i dent i c a l vowe l s , e t c . ) o c c ur o n l y i n
a handful o f ( mo s t ly non-Malay ) i t ems .
In t he dat a t here are 5 3 morpheme s c o n t a i n ing 1 1 d i fferent vowe l
s equenc e s . The fol lowing e xamp l e s ( 4 . 5 ) show t h e t yp e s of vowe l s e ­
quen c e s o c c urring i n t he 5 3 i t e m s found i n t h e dat a .

(4.5) Sequence Percentage Examples

1. i -a 24 . 5 /d i a / [ d i a ] ' he , she '
/t i ap/ [ t i ap ] ' every, each '
i -u 1.9 / t i u p/ [ t i o p] ' t o b l ow '
i -o 1.9 / r a d i o/ [ r ad i o ] ' radi o '
2. u-a 32 /tua/ [ tu a ] ' o ld '
/ b u a h/ [ b u a h ] ' fr u i t '
u-e 1.9 /kue/ [ ku e ] ' ca k e '
u-u 1.9 / s u u n / [ s u ? on ] ' r i ce n o o d l e '
3. a- i 13 . 2 /ba i k/ [ ba L ? ] ' go o d '
/a i r / [ a L r ] 'wa ter '
a-u 9.4 /m a u / [ m a u ] ' t o wan t '
/J a u h/ [ J a u h ] ' far '
66

Sequence Percentage Examples

a-e 1.9 /da e r ah/ [ d a e r a h ] ' region '
a-a 7.6 /saat/ [sa?at] ' t ime , mome n t '
/ma a f / [ m a ? a f ) m a ? a p ] ' s orry '
4. o-a 3.8 /doa/ [ doa , do ? a ] ' p ra y e r '
/soa l / [ soa l , so? a l ] ' que s t i on '

The ab ove e xamp l e s sugge st t hat t h e c ommon vow e l sequenc e s in B I are

t h o s e of h i gh and low vowe l s ( u - a , i - a ) and of l ow and high vowe l s ( a - i ,
a-u ) . Other c ommon vowe l seque n c e s whi c h are not shown b y examp l e s
( 4 . 5 ) above are seque n c e s o f t wo non-ident i c a l h i gh vowe l s ( i - u , u - i )
as i n ( 4 . 6 ) b e low .

( 4 . 6 ) 1 . I t i u p/ [ t i o p ] ' to b l ow ' (cf. 4.5 :1)

/ c i u m / [ c i om ] ' to kiss, t o sme l l '
/s i u l / [ s i 0 1 ] ' to w hi s t l e '
/ 1 i u ri [ 1 i o r ] ' s a l i va '
/ r i u h/ [ r i u h ] 'noisy '
2 . /du i t / [ d u � t , d uw � t ] 'money '
/ku i 1 / [ ku � l ] ' t emp l e '
/bu i h/ [ b u i h ] ' foam '
/ p u i l) / [ p u � l) ] , de b ri s '
/ J u i t a / [ j u i t a , J u w i ta ] ' l o ve ly '

The other o c c urring s e quen c e s are not common in t h e s e n s e that t he y are

e xemp l i fi e d only b y a very sma l l numb er o f i t ems e a c h , and most of t h e s e
i t e m s are l o an s . T h e vowe l s e quen c e a - a f o r e xamp l e w h i c h appears t o
have a re lat ive l y h i gh fre que n c y o f o c c urren c e ( 7 . 6 p e r c e nt ) i n t h e
dat a , o c curs only in a few Arab i c b orrowed word s . ( A s a mat ter o f fact ,
a l l t wo - ident i c a l vowe l seque n c e s - a -a , u - u , i - i , e - e and 0 - 0 - o c c ur
in l oanwords only , and o f t he five se quenc e s , only a - a h a s a r e l a t i v e l y
h i gh freque n c y o f o c c urrenc e whi l e t he others o c c u r o n l y i n o n e or t wo
it ems e a c h , s u c h as u - u in / s u u n / [ s u ? o n ] ' r i a e n o o d l e ' ( Ch i n e s e ? ) , i - i
in / f i i l / [ f j ? � l , p i ? i l ] ' a haraa t e r ' ( Arab i c ) , e - e in / c a m e e h / [ c a m e : h ,
c a me ? e h ] ' to be l i t t le ' ( Sundane se ) , and 0 - 0 i n / c a m o o h / [ c a mo : h ,
c am o ? o h ] ' to m o a k a t ' ( Minang ) , e t c . )
I f rule s ( c on d i t i o n s ) are t o capt ure t rue ' l ingu i s t i c a l ly s i gn i f i c ant
genera l i s at i on s ' , t ho s e fac t s mus t b e t aken into a c c ount . For t h i s
reason , s ome M S C s s e t u p b e low are marked NAT IVE t o i n d i c a t e t hat they
h o l d t rue mainly w i t h nat ive B I i t ems .
Examp l e s ( 4 . 5 : 1 - 2 ) and ( 4 . 6 ) s how t hat a h i gh vowe l may be f o l l owed
b y a low vowe l or by another h i gh vowe l . Seque n c e s of high and mid
vowe l s o c c ur only i n a few l o anwords s u c h a s i - e i n / h i e r a k i / [ h i e r a k i ]
 67
' h i e ra r c h y ' ( Du t c h/Engl i sh ) , i - o i n / b i o s ko p / [ b i o s k o p ] 'movie s ' ( Dut c h/
Engl i s h ) , / k i o s / [ k i � s ] 'kiosk ' ( Dut c h/Engl i s h ) , u - e in / k u e / [ k u e ,
kuwe] ( Ch i n e s e ? ) , /d u e l / [ d U E l ] ( Dut c h/Engl i sh ) , e t c .

[ ]
' ca k e ' ' dua l '
The c on s t raint on vowe l s that may f o l l o w a high vowe l in nat ive B I i t ems
d i s c u s s e d above c an b e s t a t e d a s fol lows :

(6) If -cons � J con

+Syll +Syll

{ ! }
+High
< aBac k >
Then [ +10W J

G::�::>J
Condi t ion : NATIVE B I

T h e c ond i t ion s t at e s t hat i f t he f i r s t vowe l o f a s e q u e n c e o f two

vowe l s is a high vowe l ( [ -C o n sonant , + S y l l ab i c , +High ] ) , t hen t h e s e c ond
w i l l b e e it he r a low vowe l ( i . e . /a/ - [ -Cons onant , + S y l l ab i c , +Low ] )
or a h i gh vowe l ; t he f i r s t and t he s e c ond ( high ) vowe l s mu s t have r e ­
ver s e d spe c i fi c at ions f o r t h e feat ure back . The cond i tion e x c ludes t he
sequen c e s i -e , i - i , u - e , and u - u whi c h o c c ur i n l o an i t ems o n l y .
Examp l e s ( 4 . 5 : 3 ) ab ove show t hat / a / may be imme d i at e l y f o l lowed b y
a h i gh vowe l ( / i / or lui ), o r b y a m i d vowe l ( /e / ) . The vowe l / a / may
a l s o be imme d i a t e l y fo l l owed by /0/ as i n / l a o s / [ l a � s ] ' g in g e r ' , / k a o k /
[ ka � ? ] ' to shout ' , etc . It ems c on t a i n i n g t he sequen c e s a - e and a - o are
very few in numb e r . The c o n s t raint o n the vowe l s t hat may imme d i at e l y

If � �
fo l l ow / a / c an b e s t a t e d a s f o l l o w s :

(6) cons
+Syll
[ � con
+Syll
+Low -Low

� +
Back

J
Then
aRound

The c o nd i t i o n s t at e s t hat if the f i r s t vowe l of a s e quenc e of two

vowe l s is spe c i fi e d a s a low vowe l ( [ -Con sonant , + S y l l ab i c , +Low ] ) and
t he s e c on d vowe l is a non-low vowe l , t hen i t must b e a round b a c k vowe l
( i . e . / u / or /0/ ) or a front non-round vowe l ( i . e . / i / or lei ) . The
c on d i t i o n b lo c k s t h e o c c urre n c e of / e / whi c h has t h e feature spe c i fi c a­
t ions [ +Bac k , -Round ] .
Examp � e s ( 4 . 5 : 4 ) s how t he o c c ur r e n c e o f t he s e qu e n c e o - a . Thi s vowe l
sequence o c c ur s o n l y in a few l o an i t ems . The i t ems / d o a / and / s o a l /
given i n ( 4 . 5 : 4 ) a b ove are b o t h Arab i c b orrowed word s . Vow e l s e que n c e s
68

c on t a i ning mid vowe l s are not c ommon i n nat ive B I word s . Of t he t hree
phono logi c a l mid vowe l s ( l e i , l a l and 10/ ) p o s i t e d in C hapter 2 , o n l y
l e i may o c cur as t h e first s e gment o f a s e quenc e o f t wo vowe l s i n nat i ve
BI i t ems . Thi s vowe l may imme d i a t e l y be f o l l owed by 101 only as in

[l
I b e ol [ b e o ] ' parro t ' , I k e o !)1 [ k e o !) ] ' s na i l ' , Ime o !) 1 [ me o !) ] ' m i aow ' ,
l
etc . The above c on s t raint c an b e s t at e d as follows :

(8) If con r- Con sl

+Syll G- Sy l Jj

J
-High
-Low
�
�::: : l
Then [ -Bac k ]

G J
d
H i gh

Condi tion : NATIVE B I

The c ond i t i on ( ru l e ) s t at e s t hat i f t h e f i r s t vowe l o f a sequence o f

t wo vowe l s i s a mid vowe l then i t must b e l e i ( [ -Cons onant , + S y l l ab i c ,
-High , -Low , -Back ] ) , and t he s e c ond wi l l be a b a c k rounded mid vow e l
( [ - C on s onant , + S y l l ab i c , +Bac k , +Round , -High ] ) . The condi tion e x c lude s
t h e seque n c e s e - i , e-e , o- i , o - a and 0 - 0 whi c h o c cur in a few l oan i t ems .
In add i t i on to the c on s t raint s on vowe l sequenc e s d e s c r ib e d above ,
BI a l s o imp o s e s c on s t ra int s on the t yp e s o f vowe l s t hat may o c c ur i n
t he l a s t t wo s y l lab l e s o f morpheme s . S i n c e j ux t a p o s e d vowe l s t hat may
o c c ur i n t h i s environment have b e e n c apt ured by t he last t hree c ondi t ions
( ru le s ) , the f o l l owing condit i on s are int ended t o c apture c o n s t r a in t s
o n t h e l a s t t wo vowe l s o f morpheme s w i t h o n e or t w o c on s onant s i n t e r­
vening in b e t we e n . The c on s t raint s d i s c u s s e d b e l o w revolve around t h e
d i s t r i b ut i o n o f mid vowe l s .
The f o l l owing e xamp l e s ( 4 . 7 ) show t hat when t he vowe l o f the final
s y l l a b l e o f a morpheme is h i gh ( I i i or l u i ) , t he vowe l o f t h e penult i ­
mate s y l l a b l e wi l l b e any vowe l e x c e p t t he front and back mid vowe l s
( le i and 10 / ) .

( 4 . 7 )la. li n i l [ i n i ] ' this '

Ip i mp i nl [ p i mp � n ] ' l ead '
It u l i sl [ t u l L S ] 'wri t e '
luJ i l [ u J i ] ' to t e s t '

l
There is one item which contains the sequence e -a namely bea [ bea] ' tax ' but this
f'orm alternates with b i a [ b i a ] and therefore it is justifiable to take the latter as
the underlying form of the morpheme . The former alternant can be accounted for by a
lowering rule ( see P-rule 2 ) .
 69
b . / s i b u k / [ s i bo ? ] 'busy '
/p i n t u / [ p i n t u ] ' do o r '
/d u d u k / [ d u d o ? ] 'sit '
/ t u t u p/ [ t u to p 1 ' shut '
2 a . / s a r i l) / [ s a r L I) ] ' often '
/ba l i / [ ba l i ] ' buy '
/ l ab i h/ [ l ab i h ] ' more '
/sad i h/ [ s ad i h ] ' s ad '
b . / s a b u t / [ s a bo t ] 'me n t i on '
/ban t u k/ [ ba n to ? ] ' s hape '
/ g am u k / [ g a mo ? ] ' fa t '
/a m b u n / [ am bo n ] ' de w '
3a. /a p i / [ a p i ] , fir e '
/m a t i / [ ma t i ] ' de a d , die '
/ r a m p i l) / [ r a m p L I) ] ' s Zim '
/nan t i / [ n a n t i ] ' later '
b. /a b u / [ a b u ] ' du s t , a s h '
/ t a k u t / [ t a ko t ] ' a fra i d '
/a l) g u k/ [ a l) g o ? ] 'nod '
/ba n t u / [ ban t u ] ' he lp '

I t mus t b e not e d however t hat t here are a few rec e nt l y b o rrowed i t ems
whi c h deviate from t he ab ove c ondit ion , s u c h a s / s e r i / [ s e r i ] ' s erie s '
( Du t c h/Engl i s h ) , /m a t e r i / [ m a t e r i ] ' m a t e ri a l s ' ( Dut c h/Engl i sh ) , / b e n d i /
[ b e: n d i ] ( Sundane se ) , / b o l u / [ b o l u ] ( Ch i n e s e ? ) . The c on ­

� �
' gi g ' ' pa s try '
s t raint i l l u s t r a t e d b y e x amp l e s ( 4 . 7 ) above c an b e s t a t e d a s f o l l ows :

(9) If i- C on sl [ +C on s ] ( [ +Con s ] ) con' ( [ +Cons ] ) +

[
l2- S y l lJ +Syll

!
+High

J
Then H i gh
-High
+Ba c k
-Round

Condition : NATIVE B I

The c ondi t i on ( ru l e ) s t at e s t hat i f the l a st vowe l of a morpheme i s

spe c i fi e d as a high vow e l ( / i / or / u / - [ -C o n s onant , + S y l l ab i c , +High ] )
then the s e c ond l a s t vowe l of t he morpheme wi l l be e ither a h igh vowe l
( [ -Con sonant , + S y l l ab i c , +High ] ) or a non-high non-round b a c k vow e l
( i . e . /a/ - [ -C on s onant , + S y l l ab i c , -High , +Ba c k , -Round , +Low ] , or / a /
- [ - Cons onant , + S y l l ab i c , -High , +Ba c k , -Round , -Low ] ) . The condi tion
e x c lud e s l o an it ems l i ke / s e r i / , /m a t e r i / , / b e n d i / , e t c . ment ioned above .
70
(4.8)1. n
(4.8)
Examp l e s b e l ow i l l u s t ra t e t hat / u / never o c curs in t he p e n u l ­
t imat e s y l l a b l e o f morpheme s when the f i n a l s y l l a b l e i s /0/ .

2. o r
/ton ton/ [ to ton ] ' to w a t c h '
/ b o d o h / [ bod o h ] ' s tupid '
/o b o r / [ b o ] ' t orc h '
/ b e s o k/ [ be s o ? ] ' tomorrow '
/ b e l) k o k / [ b e l) k o ? ] ' b e n t , curv e d '
/eko r/ [ e ko r ] ' ta i l '
3 . /Jabo l / [ J a bo l ] ' br o k e n '
/baton/ [ ba t o n ] ' concre te '
/bat ok/ [ b a t o ? ] ' (a k i nd of) fi s h '
4. /ba l ok/ [ ba l o ? ]
/ J a go/ [ J a g o ]
' ba r , b l o c k '
'cock, candi da t e '
/ r a m p o k/ [ r a m p o ? ] ' t o rob '
5. / k i l o/ [ k i l o ] ' k i lo '
/p i l on/ [ p i l o n ] ' i nnoce n t '
/ s i l) k o l) / [ s i l) k o l) ] ' ca s sa v a '

It shou ld be noted however t hat the l a s t t hree group s o f t he above

e xamp l e s are not very c ommon .

[ ]
The c on s t raint i l lustrated by examp l e s (4.8) ab ove c a n be s t a t e d a s
fo l l ow s :

( 1 0) If [ ] con + [ Con s ] ( [ +Cons ] ) cons ( [ +Con s ] ) +

r
+Sy l l +Syll
+H h -High
+Round

Then [ -Bac k ]

The c ondi t i on s t a t e s t hat i f t he vowe l o f t he l a s t s y l l a b l e o f a

morpheme i s spe c i fied as a b a c k mid vowe l ( /0 / - [ -Consonant , + S y l l ab i c ,
-High , +Round ] ) and the s e c ond l a s t vowe l as a high vowe l , t hen it must
be / i / ( [ -Consonant , + Sy l l ab i c , +High , -Ba c k ] ) .
The re i s s t i l l anot her t ype o f s e quent i a l c on s t raint whi c h c an b e
a c c ount e d for h e r e - c on s t raint s deal ing w i t h diphthongs . B I has t hree
d i p h t hongs , v i z . l a y / C a y ] , l a w/ C a w ] and / o y / C o y ] ( e qu ivalent to I PA
[ a i ] , [ a u ] and [ o i ] r e s p e c t ive l y ) . A dipht hong i s di fferent from a
sequenc e o f t wo vowe l s in t hat i t s t wo s e gment s b e l ong t o one s y l l a b l e
in t he s e n s e t hat t h e y a r e u t t ered w i t h o n e p u l s e o f s t re s s ( c f .
Hend e r s o n 1971: 92, 128; Aberc romb i e 1967: 60; 1977: 215-216,
Cat ford
whi l e the s e gment s o f a s e quen c e o f t wo vowe l s b e l ong t o t wo d i f ferent
etc . ) ,

s y l l ab l e s in t he sense that t hey invo lve t wo imp u l s e s of s t re s s i n

 71

exc itat ion . In other word s , a dipht hong has t he feat ure spe c i fi c at i o n s
[ -Con sonant , + S y l l ab i c ] [ - Consonant , - S y l l ab i c ] , w h i l e a s e q u e n c e o f
t wo vowe l s has the feature spe c i fi c at i ons [ -C on s onant , + S y l l ab i c ]
[ -Cons onant , + S y l l a b i c ] .
D i s t ribut iona l l y , B I diphthongs o c cur only in morpheme f i n a l p o s i t ion
a s in ( 4 . 9 ) b e l ow .

( 4 . 9 ) 1 . / p a n d a y / [ pa n d a y ] 'c lever '

/damay/ l d ama y ] ' pe a c e fu l '
/sampay/ [ sampa y ] ' ar r i v e , u n ti l '
/pa n t a y/ [ p a n t a y ] ' c oa s t '
/ s u rj a y / [ s u rj a y ] ' ri v e r '
2 . /kar baw/ [ ka r ba w ] ' buffa l o '
/ k a m a r a w/ [ k a m a r a w ] ' dry s ea s o n '
/ d a n a w/ [ d a n a w ] ' lake '
/ a rj k a w/ [ a rj k a w ] 'you ( sg ) ' .

/ p u l aw/ [ p u l a w ] ' i s land '

3 . /amboy/ [ amboy ] ' he y , wow ' ( in t e rj )
/s epoy/ [ s epoy ] 'breeze '
/s akoy/ [ sa koy ] ' a k i n d of w h ea t '
/ k o b o y / [ ko b o y ] ' cowboy '
/amoy/ [ amoy ] ' gi r l ( o f C h i n e s e de s c e n t ) ,

The c on s t raint s on t he t ype o f vowe l s that may b e c ome the f i r s t

(ll ) If � ]
e l ement o f a dipht hong in BI c an b e s t a t e d a s f o l l ows :

r- Consl cons

[1
li S y l lJ -Syll
+Vo i c e

-High
�

J
Then [ < -Round > ]
+Ba c k
< -Low>
< +Round>

The cond i t i on s t at e s t hat i f the final s e gment o f a morpheme i s

s pe c i fi e d a s a vo i c e d glide ( [ -Consonant , - S y l l ab i c , +Vo i c e ] ) , and t he
s e c ond l a s t s e gment as a vowe l , then it mu s t be /0/ ( [ - C o n sonant ,
+Sy l l ab i c , -Low , +Round ] ) in whi c h c a s e the f inal g l i de mus t have t he
feature s pe c i f i c at ion [ -Round ] (i.e. /y / , or / a / ( [ -Consonant , +Syl l a ­
b i c , -High , +Ba c k , +Lo w ] ) . Note t hat t h e r e a r e t hree vowe l s w h i c h
have the feat ure spe c i fi c at ions [ -H i gh , +Bac k ] , name l y /a / , /0/ a n d / e / .
The vowe l /e/ i s b lo c k e d by the feature spe c i fi c at ion [ +Round ] .
72
veryIt smusmallt number
be notedofhowever
morphemes.that the diphthong occurs only in a
/oy/

4 . 1 . 3. F I NA L S E G M E NTS

Iin atvowelwas poiornatedconsoutonant.in Chapterfurther3 thatexmorphe

A
m es
amionnation i n BI may
of BIof morpheend either
mtsesthatshows
thatmay occuBI rexhibits
i n morphea number
m e-fi n al of c o nstrai
positi o n. n ts the type s e gm e n
morphe Examples
m e-f i n ( 4 . positi
al 10) illonusaretratelimthatitedvoitocesonorant
d consonantsconsonants
that mayotheroccurthanin
t(h4e. 1pal0) 1a. tal nasal /r/ .

/ a n a m/ [ a n a m ] ' six '

/d a l a ml [ d a l a m ] ' de e p , i n s i de '
I p o h o n / [ p o h :> n ] ' tr e e '
l i k a n / [ i ka n ] ' fi s h '
/ u n d a l)/ [ u n d a l) ] ' to i n v i t e '
/ h i t u l) / [ h i t o l) ] ' t o coun t '
2./a s a l / [ a s a l ]
/Jua l / [Jua l ]
' or i g i n '
' to se L L '
/ ba t u l / [ ba to l ] ' corre c t '
/ma h a l / [ m a h a l ] ' e xpens i v e '
3./aka rl [ a ka r ]
/ba s a r/ [ ba s a r ]
'root '
' Large '
/e k o r / [ e k:> r ] ' ta i L '
/ t a l u r / [ t a l o r ] ' e gg '

pheThemeconstrai
moforrmalis fied nasalfollnpositio
ts on ntheillustrated
o ws:
type of voicbyedexamples consonants( 4 . 1that0) abovemay occancuber in
( 12 ) If G+VconsoiceJ +
The n [ +�Son]
Th e c
specifExamples o nditio
ied as ( voiced n states that if the fi n al seg m e
coshownsonant,that ofthentheitfourmustphonologicn t of a morphe m e is
be a sonorantal nascoanlsoconsonant.nants
postulated i n 4 . 1 0:
Chapter1 ) 2, onl y t h ree ma y oc c u r i n morphe m e fi n al positi o n.
Thaect,palpalatalatanasl coansl onantsneverdo notoccursoccuinr morphe
fExamples Inl
in mo r mphee mfienalfinpositi
al on. on.In
positi
( 4 . 1 1) below
may occur in morpheme final position. show that voiceless consonants other than pal a tals
 73
( 4 . 11) 1 . /ha r a p/ [ ha r a p ]
/a l) k a t / [ a l) k a t ]
' hope '
' l ift '
/un t u k/ [ u n t o ? ] ' for '
/a s a p / [ a s a p ] ' smoke '
/ i k a t / [ i ka t ] ' to tie '
l a d i k/ [ a d � ? ] ' y oung s i b l ing '
2.
/ t a r a f/ [ ta ra f , t a r a p ]
/ i n s a f/ [ i n sa f , i nsap]
' p ha s e '
' to rea L i s e '
/a t a s / [ a t a s ] ' a top '
/ b a g u s / [ ba gos ] 'nioe '
/ t ox/ [ t ox , t o h ] 'yet '

Thbyeexamples
constrain(t4 . o1n0: the1 ) anddistrib( 4 . 1u1)tionaboveof palatal
ca n be s c
t o nsonants
ated as f o illustrated
ll o ws:
( 1 3) If [ +Cons] + !
r
Then aAntaBackJ
G

featandTheularebials),
condition
specificoratistates ns eithattor,hermorpheme-f
o[ -Anteri [+Bac+Anteri o
k]ications
ir,nal-Back]consonants( i . e . musdentalst have/altheveolars
( i . e . velars). Papatals ( / � / , /�/ ,

satisfyIt wasthestatedconditionin Sec(tseioenalso Condithattion ( 1does2)). not r,occur-Back],in morpheme­

/p / ) , which have t h e feat u re s p
2.2.5.
ecif
/a/
[ - A nteri o do not
final msylle-fianbleal isyllable
morphe n nativeisBIli(mMalitaedy) toiteloanwords,
ms. The occurre s u c h nce( 4of. 12)/abelow.
as / in
( 4 . 12) / s a r am/ [ s a r am ] ' fr i g h t en i n g '((JJavanese)
a a (Dutch) avanese)( Javanese )
/ r u w a t / [ r u wa t ]
/ m a r am/ [ m r m ]
/tan ta/ [ ta n t a ]
, oomp L i o a t e d '
' to o L o s e one ' s e y e s '
' a un t '
/sen d a r/ [ s e n d a r ]

Thesyllaconstrai ' tran s m i t t e r ' ( D utch)

nnatit onvethBIe tiytepemsofcanvowelsbe statedthat mayas follows:
occur in morpheme-final
ble i n i-
( 14) If ConsLow ( [ +Cons]) + l
G J
fa
Then BackRoundJ �
�
Cond i t ion : NATIVE BI
74
Theas condi
a t i
non-l o n
o w s t
voates
w el that
then if
it the
will l a s t vowel of a morphe m e
( i is
.
k vowel front vowel
be a roundor abacnon-round e . s p
lui
eci­
orfied( i . e . - [ -orConsonant,[ -Co-nsLow,onant,+Back,-Low,+Round])
101
Iii lei - -B a c k , - R ound]). Th e vowel lal

doeswhich hasmeetthe thefeatuThenre spcondition.

not ecificationsThe[ -Consconditiononant, is-Lonlyow, +Back,true with-Round]
native BI items.
4. 2. S EG M E NTAL ' I F-THEN COND I T I ON S '

mental ' If-Thenthat Coexistnditions'betweenspeciffeatuyrethevalrangeues. ofThepossible

andvalSegudependencies y s t ate segfeatmuenrets
ment.es which follow mechanically from other features in the same seg­
4 . 2 . 1. C O NS O NA NTS

( 1 5) ThenIf [[ +-Cons]Syll ] t
All
(16) If ConSon consonants r
i
S]
n B I a r e redundantl y [ - S yllabic].
� J
Then [ �
+ V oice]
All
( 1 7) If Voice s o norantr
cons l
Cons o nants are redundantly [ + V oice].
l:- J
Then [ !
- Son]
procatesconsconsonants
tion( 1 8All) reciIfvoiceless in ( 1y6) [. -Sonorant]. This condi­
the statemaernet maderedundantl
G+NasalJ
Then �+SoV01Cn !

All nasal consonantsJ are redundantly [ +Voice, +Sonorant, -Continuant].

-Cont
 75

(19) If BacConsk [+ l
G J
�
Then Antco j- l
l.:- rJ

t(h2e0)AnreduyIfcnodantnsonantfeatwith[ureConstvalheuesfeat[u-reAnterispecific
or, -Coroationnal].[+Back] will acquire
+So-NasnalJ
Then [+CorAnt ] �

+ V oic e
Non-n. asal sonorant consonants are redundantly [ +Anterior, +Coronal,
1) If [+Soncons]
+(V2oice]
Then [+AntNlsal +Cont
+Cor+VoiceJ
The only s e g m e n t with the feat u re s p ecific a ti o ns [ + Consonant,
( 2 2) If [+Vconsoice]
and+Sonorant,
trill +Continuant]
cons o nants are is[ +theConsolateral
n ant, alveolar -Conti(nNuant].
/1/.
+Sonorant, ote that) nasal
+Cont
-Son
+
Then Antco p l
li- rJ

+ContiThenuant,only seg-Smonorant]
ent withisthethefeataluvreeolasprecific fricativatieon [ +Consonant, +Voice,
/z/ .
76
( 2 3) If [+Soncons]
The n [ +Bac
+�Nas k
a l]
There is
( 24) If [+CconsontJo n ly one bac k so n orant cons o nant, viz. In/.

The n [ +Bac
-!V k
oice] feature specif
+ConThetinuant,only seg+Bacmenkt] wiistthhethevoiceless fricativeications [ +Consonant,
Ix/ .

4 . 2. 2 . V O W E L S A NV G L I V ES

( 2 5) If l2-i-CSyllonsJl
�
Then l2-r+Vsooicn el J
All vowels
( 2 6) If [+Syllcons� a r e re d undantly [ + V oi c e, +Sonorant].
Then [+BacHigkh ] +Low
!

- R ound the featureacquiresspecifictheafeattionusre[ -valConsuoesnant,[ -Hig+Syllabic,

Theisonly vowhichwel withredundantly
+Low] lal
+B( 2a7)ck, If-Round]. [cons� h,
+Syll�gh
Then �LaBacWk ] +Hi
aRound
redundantlyand[ -Low] and always have the same specifi­
catioHinghs forvowelsthearefeatures back round .
 77

( 2 8) If [+Syllcons�
Then RouLownd f
�
l
-Back
l- J

( 2 9) If [-Syllcons ]
Front vowels a r e re d undantly [ - R ound, - L owJ.
Then �+Hilngh] + V
s oice
-Low+Cont
( 30 ) If [-Syllcons ]
Voiced glides are redunda n tly [ + Sonora n t, +Hi g h, -Low, +Conti n uantJ.
Then [+BacLk - V
s oice
+Low+ContJ
Voicelesscal gllevel,ides arethe refeatdundantly
phonologi u re s p ecif[ -Sonorant,
i cation [ + +BacContikn,uantJ+LowJ.is Oalson the
reledvelundantof repres becauseentatheretion, isviz.only one( cvoicf. Sectielessongli5. d3e. ) on. the phonological
/h/

4. 3. SUMMARY OF ' MO R P H E M E STRUCTURE COND I T I ONS ' ( MSCs )

twopreviThissecoustisecosecns.titoions.nNumbers
presentsofthcoenditisummaryons ofareconditisi m i l a ornstosetthoseup giinventhe infirstthe
(1) If liConsNasalJ [-Soncons]
r

Then �SAntCor �SAntCor 1 �

-C o nt
yBackJ yBackJ
78
( 2 ) If [+Vcan,oice� [ +Cons]
Then -Son [VOleI e]
+So-Hingh
(3) If [+Vcan,oice� [ +Cons]
-Son [VOlj e
Then +Son-HighJ
( 4 ) If -Son-C+C�ontns [ +Cons]
--CorAnt
Then -Bac[ +Voicke ] �VOice]
�
+Son
( 5 ) If [ con J -Syll
-SyllVal,, [C<+-Cons>
< J
Then [+Co+Lonwt ] Voice >]
� �

l -con, �con J
<a+HiBacgkh>J { j
( 6 ) If +Syll +Syll
Then [+Hi+Low]gh }
l -aBac k >]
Cond i t ion : NATIVE BI
 79
( 7 ) If �+Sy11
con� [+Sy11
con�
+Low �-LBacowk �

[+Sy11-Hicon]gh G+Sy11consJ
Then aRound J
( 8 ) If
Then . [ -Back] t+Round
-Low �
BaoI k � Condi tion : NATIVE BI
( 9 ) If G+Sy11consJ )[ +Cons] ([+Cons]) �+Sy11con'� ([+Cons]) +
-Hi g h
!i g h]
+High
rt:::�J NATIVE BI
Then -High
( 1 0) If �+Sy11con'� [ +Cons] ([+Cons]) [+Sy11con, ( [ +Coo>] ) +
Cond i tion :

+High -Hi+Roundgh J
Then [ -Back]1 J
( 11 ) If �+Sy11con [-Sy11con, ] +
[ ! +�
Voice
+Round>J
Then < +B_Haigckh [ < -Round>]
<-Low>
 80
( 1 2) If G cons +
Then +[ +VSon]oiceJ
�

( 1 3) TheIf n [aAnt+Cons] +
�
C-aBackJ
( 1 4) If [con J ([ +Cons]) +
-BackLow
�
Then �aRoundJ
Condition : NATIVE BI
( 15) TheIf n [[ -+CoSyllns]]
�
( 16 ) If Gcon J
Then +Son
�
[+Voice]
( 1 7) If -[L-VSon]consoiceJ
Then f

( 18 ) If Gcons
Then +Nas[VOiOa]1J
�

-C�+Sonnt
( 1 9) If Gcon J
Then +Bac[An k
�
-CorJ
 81

[+Soncon,
-Nas[AntalJ]
( 20) If

�
Then +Cor+Voice
( 2 1) If [+Soncon]
Then +C[NaM]
+
ont
++CorAnt
+Voice
( 2 2) If �con+Voice' j
+Co-Sonnt
Then GAnt
+CorJ
+

( 2 3) If [conJ
+Son+Back
t
Then [ -Voice]
( 2 4) If +Cons
+Cont
Then +Bac[ -Voice]k
+

(25) If �+SylconslJ
•
Gvoice J
Then +Son
82

(26 ) If [con+Syll'�
+[HiLowgh
•
Then +Bac-Roundk J
( 2 7 ) If [con�
+Syll
+Hi�LOWgh
*
Then aBacaRoundk J
( 28 ) If [+SyllCO"'J
Then -lRoBacundk J
*
-Low
( 2 9) If [con, ]
Then +[Vsoiceon l
-Syll
•
+Hi-Lowgh
]
+Cont
( 30) If [con,
-Syll
-VsLoice
Then +Bf+LaowckJ
+Cont
 CHAPTER 5

PHONOLOGICAL RULES

morphe Onemefundamental
s h oul d have conconeeptiandononlinygenerati one f o rmveonphonolthe phonol
ogy isogical
that eac( shyste: m­
aderiaticmovrphonemi c ) level inof various
pheed mbye undergoes
r u les f r o m the
representenviatiorn.onmentsAll orvariations
s a me underl y i n g ( p honol contexts
o gical)
iaren shapeto bewhich
f o rm. Whil e
morphe
ofpendentlwell-fmeyostrucrmefromdturemotheirrphecomndienvironme
estandionssne(Mts,gmSC'entsphonol
s ) oanrethedesigphonological
ned to speciflevely theinrange de­
arevaridesious genvinedrtoonmesnpts.ecifyInaltsernati o ns i n ogicathatl rulmorphe
shape es propermes undergo
( P-rules)in
treated u c h cases, the variati o ns are no long e r
llow frasoimnthequestio
flaonguage generalproperties lny (valc f. idofAnderspriinndiociples
vidualofmorphethe msoes,undbutstrucaretureshowofn tothe
ns) 1974:are not50-51).intendedIn ottoherdealwords,withthpar­e
photicnuollaorgiformscal rulesof inproper di v i d ual ( P -rul
morphe e m es but rat h er wi t h the s o und s t ruc ­
tureTheandP-rules sound properties
f u nctio n i n ofthethefollowi language.ng ways:
(( iii)) theythey change
operatefeatacrossures,morpheor delete, me boundaries orsensinesert,thatorcertaicoalescen pairsegs mofents
( iii) rulesthey ahavere partially to be ordered
applied i n i n the
order, other w ise the out p ut of t h e
deri v atio n will be
Basically,at a thetime.orderiIt nhasg relnoaditiornecgover i n c o rrec t f o rms.
nsntheg onoperatio n ofl organisatio
a pair of n
rules
of thewhichentirrequi e setreofnoP-rules inawhich t beari the overal
somein itheirndependent ruleson) (arei . e . placed
rules particul
or listed before others. Thus rule numbers in the following paragraphs r order applicati
83
84
dostated.not necessNor adoritheyly indicindicateateorderthe rankrelationof rulbetween rules, unless s
esmeainn thethatseitnseisthmoatreapro­o
rulductie vwithe thana smaallrulnumber does not necess a ri ' l y
ointemstheonbasiswhichofeacaspehectsrulwithetheyaoperates. biggeaccountr numberfor becaus
ratherethanthe onrulesthearenumberset upof
wordrulesTherelevelsometiarerulmtwoeess, types
(
( )
ofb transf
andalso called P-rulesormational
in generaticyclicve phonology,
phonetic rules einn the1975:literat rul e s. namelyword ( alevel)
The
ure with- seeseg­e. g .
Choms k y and
mecyclicntal aspects Miller of 1963:
phonology 3 1 4, o Fischer-J¢rgens
n the word level, a n d 2
the 45) tradeal
n sf o r matio n al
teulation
nce levels.rules dealThewithmainprosconcoedirnc asinpectsthis ofchapterphonologyhowever,on phrasis theeandform­sen­
of the f o rme r type of rul e s. The
latter only.type isAnlimatteitedmpttotoacdealcent withplacementthe trrulesansformatio discussio n of rules of the
on thenwordal cyclicand phrasrulese
levels
beyothisndisthenotphrasthe econcleevelrn woulof thed inpresvolvente thdesce discussio n of syntax, and
This chapt e r is di v i d ed i n to 11 sec t i o ns. r iptiSeco n.
t i o ns o n e through ni n e
de5. a1l0 withdealsawithnumberacceofntsegplacmentalementaspects on the of thelephonol
word vel, ogySeofction 5.Se1cti1 on
and BI .

gives the summary of rules set up in the first ten sections.

5 . 1. CONSONANT D E G E M I NAT I ON AND TR I L L D E L ET I ON

5.1. 1. C O NS O NANT V E G E M I NA T I O N

Iconsonants
t has beewinthipoin morphe nted outmesininSectiisonCC,3. 1and. thatthisthehasmaxifumrtherum sequenc
BI beenoccure of
stated 1n s
in Sisunchcesefquence.u bsectio n
This 4. 1 . 1 .
constraith a t n no
t two
extends i d entical
to the cons
phonetico nants may
level.
sonants, orematire arevesformsstems( ia. ned. words
th ( affixesconsmatyrucbegitednbywiaffith axndatiendon) inwhichcon­
)

centontaiation.n sequencSpecifice s ofally,two identical co n so n ants in the underlyi n g repres­

may( ioccur) The onsequencthe ephonolof twoogithtrills cealfolelvelowinofg trepres
wo-identical
e nt a ti o n.consonant sequences
ofas ( 5 . 1 : 1 ) or to stresemsubegiltinngnifnromg witheth prefixi
/ b a r / , /pa r - / / t a r -/
(r-r)
a trillngsuch
( i i) Theing ofsequence of two velorar stopsto stemress endiultinngginfroamvelarthe sustopffix­
/-kan/, /- kah/ / - k u/
( k- k )

such as ( 5 . 1 : 2 )
 85

( iii) ofThe/-lsequence of t w o laterals res u l (1-1)

t ing f r om the s u ffixing
a h/
( iv ) Thesuffixing to s t ems ending
sequencofe /-ofmtu/wotobilabiali n a l a teral s u ch as (5.1:3)

(5. 1 : 4)
stems endinasnaglsin a bilresabialultingnasafrlomsuctheh as
( m -m )

Notice that the

on the phonetic level. above t w o-i d entical cons o nant s e quences are degemi n ated
( 5 . 1 ) 1 . / b a r + r a n a Q / [ b a r a n a Q ] ' to swim '
/ b a r + r a t u s / [ ba r a t o s ] ' hundreds '
/ p a r + r u m a h + a n / [ p a r u m a h a n ] ' ho u s i n g s y s t em '
/ p a r + randah/ [ pa randa h ] ' to t o w e r '

/ t a r + ramay/ [ ta ramay ]
( adjectival)
/ t a r + r a s a / [ t a r a s a ] ' t o fe e t '
' t he mo s t crowded p a r t '
2 . / n a i k + ka n / [ n a i ka n ] ' t o ra i s e '
/ t u p j u k + k a n / [ t u p j u k a n ] ' to s how s o me t h i n g '
/ e n a k + k a h / [ e n a k a h ] ' i s i t de U c i o u s '
/ r u s a k + ka h / [ r u s a ka h ] ' i s i t broken '
/ b a p a k + k u / [ b a p a k u ] ' my fa t h e r '
/ a n a k + k u / [ a n a k u ] ' my s o n /daug h t e r '
3 . /amb i l + l a h/ [ amb i l a h ] ' ta k e (it) '
/ t i Q g a l + l a h / [ t i Q g a l a h ] ' s tay (p t e a s e ) '
Ijua l + l ah/ [ jua l ah ] 'seH (i t ) '
4. / k a u m + m u / [ k a u m u ] ' y o ur fo t k ( g r o up ) '
/ a y am+mu / [ a y a m u ] ' your hen '
/ m i n u m +m u / [ m i n u m u ] ' y our dri n k '

tiwnated.o-iItdmusentical
tl be noted c o however,sequenthatces arein slleongtw andheneddeliberate
nsonant rather speechdegethem­
than
thestop,sewhichquencIneisthethecai.commo
?-k,
see . ofthethefirssequenc
n realis t a tio
e stop( c isf. realisedoneasmaya glottal
velar
n of
k-k

/k / i n
5.1:2)

morpheme final
even hear
position.
rulForethis. reason, the rule set up below has to be regarded as an optional
P-rul e : Degemination
(1) + ---- �

FnwhereforCastands
specificforbuanyndleconsoof nfeatanturessegment and
1
The sequence r - r however is always pronounced as a single [ r ] .
86
Theeticallyrule realis says theatd asa asequenc singlee ofconstwoonant.identical consonants may be phon­
5 . 1. 2 . TR I L L V E L E T I O N

Here we are goi n g to deal

prefthe isexesquence ( resulantding from theWe sprefiaw inxinexamples
/ba r / , /pa r/ /tar/.
with the' d el e ti o n of t h e
(5.1:1)
fi n al
/r/ i n
abovetothat the
r-r g of thos e prefixes
igestenmmnoinsrationmalbeginsnipruleech,nge with( Pand-rulathisetrill)1) pheno
s e t
ismphonetically
up enon
above. is perf
Th e e realis
ctly
rul e
ed uasreda bysinthegle de­
capt
does not i n dicate
[r]

or not.A furththeerfirstexaminseatigmoentn of(orthethephonetic

whether second) of the sequence is deleted
s h
r-r

apes of those pref i xes s h ows

thatfixesthereas beiinsgadeleted good reasratonherforthanregardidegemngintheatedfiwhennal theyincomthosebine withpre­
/r/

stems beginare,ningrespecti
andtrill.
/ t a r/
with a trill. v ely, That is,aswe may regardandthat before a
realised [ ba ] , C p a ] eta]
/ba r/ , /par/

This andfact has beenare reflected i n the orthographi c syst e m i n that

/b a r / , / p a r/ /tar/
stesounds.ms begiInnniotherng wiwords, th a trilltheresandfoprmsectiasvely speltandasand are before
ber - , per-
be- , pe-
be- , pe-

te-
ter-
and otherbefore
consi
te-

deredpre­to
befixtesheandcorrecstemtsfbegiormsnofningsyllwithablesa triwheln arewordsdivconsisti i d ed i n n
tog of those
syllabl e s, s u c h
as be r a t u s
t e ra s a
(be-ra-tus)
( te-ra-sa)
' hundreds ' ,
' t o fe e l '
perendah ( pe - re n - d a h )
(meadjnetctidiscussed ' to lower ' ,
val) ( c f.above,5. 1 : 1the) . forms
Iand n addition
te- (oconsists to the envi
r of a andconsonant alsoplus occur
[ ba ] , C pa ] eta])
r o n be- , pe­
befomiredstevowmels whosplues afirtrillst
sylla b l e
as( 5 . i2n) 1 (. 5 . 2 ) below. a c e ntral
b e k e r j a / b a r + k a r j a / [ b a k a r j a ] ' to work '
b e s e r t a / b a r + s a r t a / [ b a s a r t a ] ' toge t h e r , p a r t i c ip a t e '
b e t e r n a k / b a r + t a r n a k / [ b a t a r n a ? ] ' to ranc h '
2.p e k e r j a a n / p a r + k a r j a +a n / [ p a k a r j a ? a n ] ' wo r k , o c c upa t i o n '
p e s e r t a a n / p a r + s a r t a + a n / [ p a s a r t a ? a n ] ' p a r t i c i pa t i on '
p e t e r n a k a n / p a r + t a r n a k + a n / [ p a t a r n a k a n ] ' ranch '
3.t e k a r j a k a n / t a r + k a r j a + k a n / [ t a k a r j a k a n ] ' mana g e d to do '
t e s e r t a k a n / t a r + s a r t a + k a n / [ t a s a r t a k a n ] ' i n c l uded/s e n t u n i n t e n ­
t i o na l ly '
t e t e r b a n g k a n / t a r + t a r b a Q + k a n / [ t a t a r b a Q k a n ] ' manage d to f l y
some t h i n g '
 87
wioftNotic
hthea cfioensnthatalonant.theinAlsecothoughnd syllP-rulablandee ofhasthecaptabovebefuredoresteainmtrill,
/r/ /ba r / , /pa r /
1
/tar/
saalwawyaysthebeginsdeletion
this phen­
[+Son-Nascon,alr-� CVa --- + [+Son-Nasalcon, ]
omenon is also accounted for by P-rule 2 below.
P-ru l e : Tr i l l deletion
(2) / ,

-Cont -C-Conson [con, ] X

C +Syll
+Back
-LRowound -Cont
+Son-Nasal C
where C V stands for the firs t t w -
o seg m e n ts of the
prefixes
begi a
/ba r / , /pa r/
/r/, X
a n d
/tar/,
othernnithanng with aandvowelfor any sequence of segments C for a y co n sonant
Th e rul e states t
-Csteontims nbegiuantJ)nninginwiththe prefih at the fi n al
/r/ ([ + Cons
xes or before stemansdwhose fisirstdeleted
/ba r / , /pa r/
o nant,
/ t a r/
+Sonorant, -Nas a
before l,
begivowelns with-a[ -consonant a trill,
other than a trill followed by a syll
central a ble mi d
a trill. Consonant, +Syllabic, +Back , -Round, -LowJ) and ends in
( / a/

5. 2 . G L OT T A L STOP INSERTION , AND G LOTTAL STOP R EA L I SAT I O N OF / k/

5 . 2 . 1. G L O TTA L S T O P I NS E R T I O N

voalsowelsIt shaseatn beentihnesmaxiupOinted

mbsectiumotonout4.occurJi. n2 . Secthatintaionsequenc
3.1.
sequencthateeswithin
of
BI allmorpheows monles.y twWeo have
t w o i d entical vowels
are(mospossible
t ly) non-Malalthough
a y origi theyn . occurThis kionlndy ofin asequenc smallenumber
s ( i . e . ofsequenc
itemseofs of
twocase,idaentical vowels) is not
glottalnt stopis true notisonlinsy withi
([?J)
per m itted
ertednbetweenon the phonetic level.
thebuttwoalsoidentical I n suc h
vowels. a
Thiseme boundaries
constrai as illustrated by the following examples morph­ morphe m es (5.3) .
ac r oss
( 5 . 3 ) 1 . / s a a t / [ s a ? a t ] ' t i m e , mome n t '
/ t a a t / [ t a ? a t J ' ob e d i e n t '
/ m a a f / [ m a ? a f , m a ? a p ] ' so rry '
88
/ n a a s / [ n a ? a s ] ' un l uc k y day '
/ s u u n / [ s u ? o n ] ' ri c e n o o d l e '
2. / k a + a m p a t / [ k a ? a m p a t ] ' the four t h '
/ s a + e k o r / [ s e ? e k : H ] ' one ' (animal classifier)
/ d i + i s a p / [ d i ? i s a p ] ' to b e s m o k e d/ s u c k e d '
/d i + i ka t / [ d i ? i ka t ] ' to be tied '
/ b a c a + a n / [ b a c a ? a n ] ' reading ma t e r i a l '

Notice t h at the juxt a pos e d vo w els i n

/sa+ekor/ above are not qui t e the
bothsame.ThehaveTheyphenothmeagreeenonfeatuonlydiscre uspossednecifictheaboveafeattiounscaresn[be-High,formul-Loaw,tedand-RasoundJ.follo( ws:i . e . they
high , low round

( 3 ) [-Syllcon' l / +Syll
P-rul e : Glottal stop insertion

0 - - - - -+-
-Cons -Cons +Syll
--ContVoice aHiBLogwh ( + ) aHigh --

B L ow
The rul e s a ys that a glottal yRound stop ( [ ?J - [ - C yRound
onsonant, -Syllabic,
-tVheoice,same-Conti s p nuantJ)onsmusftorbetheinsfeatertuedresbetween two voandwels which have
ecificati high , low round .
shouldbybeexnotamplesed how(ever5 . 3 : 2that) areinofrapiten dreducspeech,ed invowel
illbutuItthisstrated to si n sequenc
gl e es
vowels,
kind of pronunciation is considered ' sub-standard'.
5. 2.2. G L O TTA L S T O P R E A L I S AT I O N O F /k/

islevel.notIt hasnecess beenaryarguefordthein repres subsectentionati2.on1 . of1 . morphe that themesglonottalthe phonol stop ( o[ gic?J) al
I t wa s poi n ted out that
( in natiyinveg andconsosnantome loan itLetems) usisnowsimplyobservea phoneti [?J occurri n g i n morphe m e fi n al positi
alternantconcofernitheng
cexamples o n
underl
thein eacrealisation /k/
/k/ .
ofumber) iinn morphe s o m e
mlleowifinngalexapositimpleson.( 5 . Notice thatthe itesamems
h
s(t5e.ms.4 )la. group ( n the f o 4 ) have
/ g a r a k / [ g a r a ? ] ' t o move '
b.c. / g a r a k+ l a h / [ g a r a ? l a h , g a r a k l a h ] ' mo v e i t '
/ g a r a k + a n / [ g a r a k a n ] ' m o v e me n t '
2a.b. / s o r a k / [ s o r a ? ] ' to s h o u t '
/ s o r a k + r a / [ s o r a ? r a , s o r a k r a ] ' t he way he s h o u t e d '
c. / s o r a k + i / [ s o r a k i ] ' to s h o u t a t ( s ome one ) '
 89
3a.b. / b a i k / [ b a j ? ] ' go o d '
/ b a i k + l a h / [ ba j ? l a h , b a i k l a h ] ' a Z Z right '

4b.ac..
/ k a + b a i k + a n / [ k a b a i k a n ] ' k i ndn e s s '
/ p e n d e k / [ p e n d e ? ] ' s ho r t '
/ p e n d e k + J1 a / [ p a n d e ? J1 a ] ' i n s ho r t '

5a.b.c.
/ k a + p e n d e k + a n / [ k a p a n d e k a n ] ' ab br e v i a t i on '
/ d u d u k / [ d u d o ? ] ' to s i t '
/ d u d u k - k a n / [ d u d u ? k a n , d u d u k a n ] ' to s e a t ( s omeone ) '
c. / d u d u k + i / [ d u d u k i ] ' to s i t on ( s ome t h i n g ) '

Theis phonetically
above examplesrealised show thatas a/k/glottalin stestopm final positias aonglottal ( c f. a-items)
aasvela avelr astopr stop when whenit isitfollowed
[kJ by a([?J ) ,
c o nsonant ( c f. b-ite m stops) andor
([kJ)
Notice (Tthathe ltheattersequenc ek-k ( isf. followe
c 5. 4 : 5 b d) mayby abevowelrealised( c f. asc-items).or
[ ?kJ

tionInruladditie ). on to therealisenviatiroonnmhasentsbeediscn ucassedpturedabove,by P-rulmaye also- degebemina­

[kJ.

/k/
1

realised
items suchasas in morpheme medial position in a small number ofetc.loan
[?J
/ l akna t/ [ l a?natJ ' cu r s e ' , / m a k n a / [ m a ? n a J ' imp l i c a t i on ,
meaning ' , /yakn i / [ ya ? n i J ' name ly ' , / r a k y a t / [ r a ? y a t J ' p e o p le ' ,
(/k/ in morpheme-medial position is usually realised ( aasdv), such as
IJaksa/ [ Ja ksa J ' j udge ' , / l aksana/ [ l aksana J ' like '
[kJ
/akrap/
[akrapJ ' i n timate ' ,
ofabove,exceptions to the l i k l i m/ [ i k l L m J
reg u l a rity ' c l im a t e ' ,
observed i n etc. ) i. n Theexamplesother (type5 . 4 )
a-items
is the real i s a ti o n of the underl y ing/k/ as a velar stop ( [kJ)

ofrathloeranwords,
[ l akJ
than a suchglottalas stop in morpheme-final position in a number
' to sea l ' ,
([?J)
/pak/ [ pa k J
/ fana t i k/ [ fa n a u k J
' p a c kage ' , / r a k / [ r a k J ' s he l f ' ,
' fa n a t i c ' , / a p o t i k / [ a p o t L k J
/ 1 ak/

' chemis t ' ,

toOn bethemarked etc.' +P-rulItemes4'continatihneinglexiconinwhemorphen me-imsedialrealisedpositiason have
/k/
/k/ [?J.
ot h er hand, i t e
theThelexicfoollnowhenwing ruleis( Prealised
/k/
m s endi n g i n
/k/ have to be marke d
asaccountsratherfor thethanrealisation of as
[kJ [?J.
' - P-rul e 4' in
[?J in morpheme-final position. - r ul e 4) /k/

P - ru l e : Glottal s top rea l i sation of / k /

( 4 ) +Cons -S-Vyoiclle -Cons

-SyllVoice ----++ ( CVC+)
+-LBowack �
-
+Bac+Lowk /

-Cont
where C stands for any -Cont
consonant and V vor any vowel
 90
rule says thatis phonetic
The+ContinuantJ) /k/ ([a+Consonant,
lly realis e -Syllabic,
d a s [?J ([ -Low,
--VConsoiceonant,, +Back,-Syllabic,
km,e+Low,ending-Cointin nuantJ)is follinomorphe
whe-wiVoice,tnh a moco+Bacnsrpheonant, /k/ wed by me-final
a s u ffix position.
( C V C + ) begiAndnning
aoptigloonal.ttal stop. Thetherealisatio
a final /k/nmayof stillinbethepholatter
/k/
neticallyenvironme realisednt isas
5. 3 . D E L E T I ON OF / h / , SY LLABLE R E DU C T I ON , VO I C I NG OF /h/ AND PLACE
ADJUSTMENT OF / h/

5 . 3.1. D E L ET I ON O F /h/

spronunci dtanceredes. frManytoicabetivesusipeakers(n/gh/s) uinb-standard

peech,Speakerswiatithouton,of BIunderthetenriskdcerttoaofdropinbeicircnthguemsconsiglottal tend
their
to
delete
trated /h/
by exa whemnplesthis( 5s. e5g)mbelentow.occurs in morpheme final position as illus­
( 5 . 5 ) 2.1 .
/ l a t i h/ [ 1 a t i J
' t i re d '
/ b a s a h / [ b a 5 a J ' we t '
3 . / l e l e h / [ 1 e 1 e J ' to tri c k l e , m e Z t '

5.When the above ( stem)J morphemes combine with suffixes beginning with
4. /panuh/ [ panu ' fu l l '
/ k a s i h / [ k a 5 i ] ' to g i v e , to l o v e '

onsonants, bytheexafinalmples ( 5of. 6 )thosbeloews. temNotice

cillustrated /h/ s tendsthatto bethedelfinaletedsegstmillentas/h/
isnuffix' in ' theinterrogati
/ - ka h /
followi suffix'esandalso/-lteanh/ds ' itmoperati
vneg exampl be ve/edmphiaticn thefosarmimeng
droppe
way( 5 . 6a) s1 . that of the stems cited above.
l 1 a t i h + k a h / [ l a t i k a J ' (are y o u ) t i r e d ? '
2.
/ b a s a h + k a n / [ b a s a k a n ] ' to we t '
3 . l 1 e l eh+kan/ [ l e l ekan ]
( causative)
' to tri c k Z e s o me t h i n g '
4. / p a n u h + l a h / [ p a n u l a ] ' ( i t wa s ) fu l l '
5.When the stems given in ( 5 . 5 ) above combine with suffixes beginning
/ k a s i h +p a / [ k a s i p a ] ' h i s/ h e r l o v e '

withor vowels'(abstrac
I-ani
( i . e . tI-i) noun/ ' transitive/c a us a tive/l
formiexcngeptsuffix'when) t,hetheprecedio c a tive formi n g s u ff i
finalng vowelof thos( i . ee . stthemse
/h/
x'
nayvoweltendof tothebeladropped s t syll a still,
bl e of the ste m i n question) is i d entical with
immediately Thusfollowinginthestefimnalfinal/h/positi( i . e . ontheinvowelexamplesof the( 5 . 7 )
tshueffixvowelin question). /h/
may optionally be deleted but not in examples ( 5 . 8 ).
 91
(5 . 7 )1 . / k a + L J T I H + a n / [ k a l a t i a n ] ' s ta t e of b e i n g t i r e d '
2 . / m n + B A S AH+ i / [ mam b a s a i ] ' to w e t ( s ome t h i n g ) '
3.
/ L E L E H + i / [ l e l e i ] ' to t ri c k Z, e o n t o '
4 . / p a N U H + i / [ p a n u i ] ' t o fu Z, fi Z, '
5.
/ KA S I H + a n / [ k a s i a n ] ' (wha t a ) p i ty '

Th( 5e. 8fi) 1n. al of stems in the following examples is never deleted.
/h/

/ k a + B A S A H +a n / [ k a b a s a h a n ] ' s t a t e of b e i n g we t '
2.
/ KA S I H + i / [ k a s i h i ] ' to z' o v e s o m e b o dy '
3 . l L a B I H + i / [ l a b i h i ] ' to i n c re a s e , g i v e more '
4.
/ k a + M A R A H + a n / [ k a m a r a h a n ] ' a n gr i n e s s '
5 . / k a + S U S A H + a n / [ k a s u s a h a n ] ' s adn e s s '

Apart from thewithin

intervocalic /h/
enviromorphe
nmentsmesillualsostratedtendsbytoexabemplesdropped( 5 . 7when) above,it
servesandas adjecti
verbs the onsvetesofunderthe thefinalfollsyllowianbleg conditions: of morphemes, especially in
(( ii)i ) VThle ( pfinalrecedisyllng avowel)ble doesandnotV2 end( folloinwaingliquidvowel)oraretrillnotoridentic
gli d e;a l;
( iii) Vhigl mush ort abe loawlovowel.w or a high vowel and V must be respectively a
2

Ideletedntervocon athelicphoneticin thelevel.followiNoticeng examples

/h/ ( 5 . 9 ) ma y optionally be
thata/) andin thefollofirswedtbyfoura hiiteghmvowel
s
bel( i . oew. oirs precededandbyinathelowlvowel
/h/ ( /
(( /5 .i 9/)) la. nd followed by a low vowelast ( /itea/).m is preceded by a high vowel
/u/ /i/) ,

/ t a h u / [ t a u J ' to know '

/h/

2.3.
I s a h u t l [ s a o t ] ' t o re s p o n d '
! J a h i t l [ J a L t ] ' to s e w '
4 . Ipah i t / [ pa L t ] ' b i t ter '
/ l i ha t / [ 1 i a t ] ' to see '
5.When the above morphemes ( stems) combine with affixes, the glottal
belfricati
o( 5 . 1w.0) 1v. e still tends to be deleted as shown by examples ( 5 . 10)
( /h / )

/ k a+TA H U + a n / [ k a t a u Y a n ] ' to be d i s c o v e re d/known '

2.3.
/ b a r + S A H U T + a n / [ b a r s a u t a n ] ' to r e sp o n d ( r e c i p r o ca Z, z, y ) '
/ p a +m a n + J A H I T / [ p a p J a L t ] ' ta i Z, o r '
4 . / P A H I T + k a h / [ p a i t k a ] ' ( i s i t ) b i t te r '
5.
/ L I H A T + l a h / [ l i a t l a ] ' ha v e a Z, o o k '
92

/h/
Atinfirstthe abovesightsteonemsmayis beconditemtpiotneded tobyconclthe presude thatence theof thedeletion dental/ of
alveolar
stems orratbystoherpthethanabsenctheserviefofanctganyasthattheconstheocodanantaboveofinthesthetemlastsfiarenalsylleither
(/t/)
syllablaeblverbsofe (those
c f.or
5.9:1)
adjectives. Items like etc. clearly refute such conclusion because
Itahul [ tahu ] ' b ean c a k e ' ,
Isah i tl [sah � t ] 'mar t y r ' , I w a h i t l [ wa h � t ] ' one ' ,

/h/
/h/
inn nouns
i these iistemnevers is never
deleted. dropped.
l There Generally
are howe v speakithreeng,nounsintervocalic
er to my
knowl
dropped edge,by mostcontaisnpeakers
ing usiinngthe' slasttandard'syllapronunciati
/h/ ble where oitn teofnBI.ds toTheybe
areIn examples we find no between and /a/ ( c f. condition
I t a h u n l [ t a Q n ] ' y e ar ' , I t a h i l [ t a i ]
(5.9) /h/
'dirt ' , I p a r a h u l [ pa r a u ]
/u/
'boat ' .

iii).itous inThetheabsencsensee thatof itthere

ems contaiis noniverbng oriadjective
/h/ n that enviro(tonmemyntknowledg is fortu­e )
contai
containning suoccurri
/h/
sequence . Thereand aresucha fewasnouns however which
ch soundng between /u/ /a / I t uhanl [ t u ha n ] ' Go d ' ,
I l uhakl [ l uha ? ]
deletedThe underl etc. Theon'./h/ segment in such items is never
' di s t r i c t ' ,
in BIyi' nSgtandardinPtheronunciati
/h/ f o llowi n g exa m ples is never deleted bec a use
endiit isngpr-cin eadedliquibyda, mitrilld voworel ( p( alc f.atal) glideor ( citf. occurs in morphemes
( 5 . 1 1 ) 1 . I s a h a t l [ s a h a t ] ' he a l t hy '
5 . 11 : 1 )
5 . 11 : 2 ) .

I d a h a m l [ d a h a m ] ' t o cough '

I l o h a l [ l o h a ] ' t o say morning p r a y e r '
2 . l 1 a h i r l [ l a h L r ] ' b orn '
I m a h i r l [ m a h L r ] ' fl u e n t '
/ J a h i l l [ J a h � l ] ' i l l - he a r t e d , s t up i d '
I l i h a y l [ l i h a y ] ' cunning '

Excludinwhichg themusthreet benounsmarkedmen' t+iP-onedruleabovein the lexicon, theanddeletion

Ipara hu/ ) 5'
( / tahun/ , Itah i l

ofabove cainn bethefoenviro

/h/ n ments
rmulated as follows: ill u strated by examples through (5 .5) (5 . 10 )

l
Some regional pronunc iations ( e . g . South Sumatra accent - see Halim 1974 : 17 8 ) also
tend to drop /h/ in nouns like /dah i / 'forehead ' , Ibahul 'shoulder ' , etc .
 93

[-S+Lowconsyll�
P-rule : Deletion of / h / ( o pti o nal)
l-Cl.Cl.+SyllLHowigh [Cl.Cl.+SyllLHowigh ( [::���
(5) - - - - -� 0 / <Va>
]) ++
J J
+Cont -

� +Nasal> Verb
-Cons -Cons
wheVa rande CVbstandsare diffforerentany consonants and V for any vowels Adj
may( a )Theoptiruloocnallyecurssaysbeinthatmorphe
/h/
deletedmei(nfi[ -ntheConsal opositi
/h/
fnant,
ollowionn-Syllabic,
environmeess+Low,nts:of the+Contisuffinuant])
gregardl x
(eingmethnet cisasenot,of bystemthes) added
svowels; suffixatito oin,t tosutherroundedextentbythattwo theidentical /h/

( b ) tivesoccursand sbetween
/h/ two non-i
ervesvowasels,theregardl d entical vowels
onseteofss theof theirfinalpositi i n verbs a n d adjec­
syllableon wiofththerespecitemst
where /h/
t h e two
to-Low]theand [ -seHigmgh,ent,+Lhaveow]. theThe featrulueredoesspecificati o ns [ + Hi g h,
in non-nasal sonorant consonantal (non-syllabic) not applystoegmeintets.ms ending
5 . 3 . 2. S Y L L A B L E R EV U CT I O N

isylls theClosely
cableasoferelpolysyllabic
ofatedintervocalic
to the phenomesDearvidiscussed
/h/ n g as the inonsseubsection
t of 5.3.1.
the second above
usconuallyditions.deleted together withitemsthe( forprecmatiedives).ng vowelThisunderintervocalic the followingis
/h/

(( iii)) ThThee vo(pwolysyllael of t b

h ic)
e fi i
r t
s em
t syll contai a n
ble s ofn o clos
the e
ited m syll
i n a ble.
ques t ion lsi .
is
As a resuthleret ofarethequidelteetiaonumber
vowel), n of intofervocdoublalicets in BI,and that (tishewordsprecediwhicngh
/h/ /a/

havebelowt. wo phonetic forms each - full and reduced forms, such as ( 5 . 12 )

94
( 5. 2) 2.l .
1 / b a h a s a / [ b a h a s a ] , [ b a s a ] , language '
/ c a ha y a / [ c a h a y a ] , [ caya ] ' light '
3.4. / s a h a J a / [ s a h a J a ] , [ s a J a ] 'jus t '
/ b a h a r u / [ b a h a r u ] , [ b a r u ] 'new '

6.5. / d a h u l u / [ d a h u l u ] , [ d u l u ] ' forme r l y , e a r l i e r '

/ d a h i n a / [ d a h i n a ] , [ d i n a ] ' m i s e ra b le '
7 . / b a h a g i a / [ b a h a g i a ] , [ b a g i a ] ' happy '
8. / r a h a s i a / [ r a h a s i a ] , [ r a s i a ] ' s e c re t '

ThosThusealdoublong ewithts arthee alfrormseady reflected in the current writing system.

baha s a , cahaya , sahaj a , baha r u , dahu l u ,
da h i na , b a hag i a , and etc.and one alsorespectifindsvelfoyrms.
rahas i a ,
� a j a , ba r u , d u l u , d i n a , bag i a ras i a
basa , caya ,

Th e doubl
toto beseekreasonabl e t
a phonologic phenome n on i s s o regul
allynableplausoniblhistorica r that i t
e explanatial lionnguistic s e e m s nec e ss a ry
to it. groSomuends.pairThuss seemto try
bahasa/basa
y explai
theto thpaie rfact that the variants are( ctf.he5.products
' language ' 12: 1 ) forofexatwmople,diffecarentn bedevel­ ascribed
opmentsm 1974:of the28-29).same SincsourceethenamelysourcSanse itekrimtcontains a sound se( quenc
Eali�hich b h i s i ' l anguage ' see e.eg .
rrsapdeeakerssdoesomesmodificati
not occuronsin Malto aity . ( BI)In ,makiit nisg natsucuhralmodificthatatiospeakers
eemedof developme
to be dinvtidhappened
ed into ttowothegroups,sourcande itesmubs. equently, n s, the of BI
twodirec­
direc t ions
tigivenon orhere)processto yieldneedsthefourforrulm es (not necess a rily appli e Th
d e
in firs
the t order
insdo notertionoccurulr eintoBI,break( 2 ) upderetrofl bahasa ,
the sequenc viz.e ( 1bec) voawusele as( inpirthisatedcaplosise ves
[ bh ]
/a/)

intoIPAthe laminino-alto veolarand ( 4 ) (vo3 )wexionvoelwellorulwerishortentogenirulchange

( = [a : J ) [aJ ,
[sJ, nge rulto etchange
htoe retchangeroflex
[ J
[sJ
[aJ

as( derifovllows:ed by the first rule above) into The process can be outlined
Forms Rules
[a] .

* b hi � i
* bahi�i
* bahisi
ScDehrwaetroflexi
insertioonn
* bahasa
bahasa ( [ bahasa J )
Vowel
Vo w el short
l o weri eningng
fJrmThebecaussecondeviz.BIdirecti
[bJ
ba s a , (1)
on p(pirrocess)
deas ation requie toreschangethreetherulasesptoiratedyield theinto
rul
does not have aspirated plosives, ( 2 ) deretroflexion [bhJ
 95

rulvowele toshchange t h e retrofl [sJ e x i n to the l a mi n o-al v eol a

[sJ, r a n d
(3)

lined asortfoellnionws.g rul(Tehetoruleschangemight notintoapply inThetheprocess

Form
[aJ

Rul e s
[aJ .
order maygivenbehere.)
out­
* b ha � a
* ba � a
* ba s a
Deaspiration
Deretrofl exieoninng
basa ( [ ba sa J )

A simi(dlaerived r explanati Vo w el s h ort

oSansn willkrit also be trueThewithformthe pair may be ca haya/caya
' t i gh t '
descri b ed as devel f r
o om
pi n g *chahya ) .
f
chahya r o m through cahaya
the followi n g phas e s. ( T he
rules might not apply in the order given below.)
Form Rul e s
*chahya
*cahya
*cah ya
Deaspiratio
Vowwelel loweri n
( schwa)nginsertion
cahaya ( [ cahaya J )

Andthe tfhoellowiformng phases.may be(Adescgainr, ibedthe rules

cava
Vo
as develmightopinnotg frapply
om in ththrough
�hahya

given below.) e order

Form Rul e s
*chahya
*ca hya
cava ( [ caya J )
Deaspiration
/h / -deletion
Howerestver,of thethispairsquasi-cithedistoricin exaalmplaccount does
eWes could perhaps
( 5 . 12 )
not seem to explain the
above andassumanyme thatothersthewhich
aresyllnotableSansreductikritonborrowed proc e ss ite m s.
through /h / -del e ti o n i n these words occurred
muchwas bylater,analogyi. ewith. aftether theSansborrowi k ri t f normsg ofththeat native
Sanskritspeakers
items, ofandBIthbeganat it
towhetherreducthise thewassylltheablescase.of polysyllabic items. It may be doubted
Another possibl
iFitraccounts for both e explborroweanatidoandn, whichnativeseBIemsitemorems mayacceptable run as because
follows.
estgl,ottalunderthe /hfrithe/-delcawitivdee.tioespnreadForrulexample,
ththrough tendency theto dropitem/h/, speakers of( c Bf.I deleted
sahaja ' j us t ' 5 . 12 : 3 )
d becomeed by a furthThen,er rulthee intwo­to a
wdwoulbe reduc*saaj a .
isidnengltical- v owel s e quence coul
e vowel resulting in ( In the case of forms containing
saj a .
96

ssucequench ases of two non-identic( c f. al vowels deria ruleveddeleti

*dau l u , *da i na 5 . 12 : 5 , 6 )
by theng/hthe/-delfiretistonvowelrule
of thosdeletio
vowel e sequencn ) ecouls would bed bseaidrequito rbeed.)subjectThe votowelthedegefacmtinthatationpoly­( or
syllabicIn theitepresmsent( stedescms) rareiptiolessn, thcoemmosylln ainblBI.e reduction phenomenon dis­
cussed treatede rulase (aorsipartngleofphonetic
shortnoeniseisparat
tvohwereel siabove a rule) in the sensfore tthhate
eventthat accounts
sylldiacahrblonice reduction ng/vowelis videleti
e wed other words,than athe
oans desca syncribhedronicabove.processIn rather
one.
It mustthebesyllnotedablehowever there alarethougha fewtheyitemmeets whichthedoconditi
thatn process not ons
undergo reductio
pointed outetc.above,Thessuechiteasms must be marked ' -P-rule in the lexicon.
dahaga [ d a haga ] ' t h i r8 ty ' , pah a l a [ pa h a l a ]
' re ward ' , 6'
The syllable reduction process( optimaoynal)be formulated as follows.
+ C [+Syllcons]
P-ru l e : Syl lable reduc tion

+Low �+-C-L�oownt��
(6) v C V (C ++
V)

------
2

�
3

1 2 0 0 5 6 7 8 9 10
4 5 6 7 8 9 10

Thewhere rul e C s stands

a ys that f o r
t h any
e vo c
w o
elnso n
of ants
the and
fi r s V
t for
syll any
a bl e vo w
and els
the ons e t of
tfollhe osecond
wi n g syllatblions:e of a polysyllabic item may be deleted under the
condi
vowel of th+Syle filarbic,st syl+Llaow])ble of the item is /a/
( a ) Th( [e-Consonant,
( b ) Th( [e-Consonseontant,of the-Sylsecond
l a bi c , syll
+ L o a
w bl
, e of
+Conti the
n i
uant])t em is /h /
e ends in a iteconmsonant).
(c) Thesylla(pblolysyllabic) contains no closed syllab le ( i . e . a
5. 3. 3 . V O I C I NG O F /h/

speakersunderofthBIe citerndcumstanc
notsecManytdeleted to pronounce s pOi eteindteoutrvocalic
n in the /h/last(whentwo itsub­is
ions) with vocal cords in vibration, as in examples below. ( 5 . 13 )
 97

( 5 . 1 3 ) 1 . I p o h o n l [ p o h :> n , p o n :> n ]] ' tr e e '

l 1 ehe r l [ l e he r , l ener 'neck '
l a h a t l [ a h a t , a n a t ] ' Sunday '
I s i h i r l [ s i h � r , s i n L r ] ' b l a c k ma g i c '
I s u h u l [ s u'h u , s u n u ] ' t emp e ra ture '
I k a h a n d a k l [ k a h a n d a ? , k a n a n d a ? ] ' i n te n t i o n , wi l l '
2 . l 1 a b i h+ i l [ l ab i h i , l a b i n i ] ' to ma k e more '
I k a s i h + i l [ k a s i h i , k a s i n i ] ' to l o v e s o me b o dy '
I t a m b a h + a n l [ t a m b a h a n , t a m b a n a n ] ' addi t i on '
I p a r + r u m a h + a n l [ p a r u m a h a n , p a r u m a n a n ] ' ho u s i n g '

The voici
( 5 . 13 )
n g of
/h/ in i n t e rvocalic positi o n ill u s t rated by examples
tfhuencfeattion.aboveureNorhasdoesnoforliitnthguistic
voice eindicglaotet asiganyl nfrificpartic
icatiancveeulawhatsr speecoeverhash novarieties
( /h/ )
incontras
the sensetivethat
or
dilowalelevelcts. ruAccordi 7

P-rule : Voic ing of / h /

le withnglrespec y , P-rtultoe thesewhole
( optional)
t up belsetowofhasrulesto beofregarded BI as a
phonology.
[-Voice] [ +Voice] V -Cons
- - - -� /
++ContLow ( + ) V
-Syll
where V stands f o r a n y vo w els
-iVtoicThisee f])orulllomaeywedsabecoysbymthatea voicgrammatic
/h/
ed ([in-aConsoilntfeonrmativ
rvocant,alice-Syllabic,
posi t ion +Low,regardless
boundary ( + ) or not.
+Continofuant,whether
5 . 3.4. P L A C E AVJUSTM ENT O F /h/

soccounduBypiesproduc
definitioed nby, tlohweeriunderlyi
i n neut r al positi n g o the
n.
nbodyg ofisthe[ +Lotongue
/h/

St u dies i n
w] becausbelowe ittheislevela glottalit
phonetics show that
/h/ dis­
plays
of whetherperfethect co-artic vowel isula[+tioLow]n withor [the+High].vowelThusthatSweetfollowswritesit regardless
H i s . . . l i ab l e t o h a v e i t s c h a r a c t e r m o d i f i e d by t h e c o n ­
f i gur a t i o n o f t h e mout h ; a n d t he p o s i t i o n f o r t he v o w e l wh i c h
f o l l o w s t h e H b e i n g g e n e r a l l y a s sume d , o r a t l e a s t p r e p a r e d ,
w h i l e t h e H i s b e i n g f o r m e d , t h e H n at u r a l l y a s sume s t he
c h a r a c t er o f t ha t v o w e l .
( Henderson, ed. 1971 : 15 4 ) 1

I
This phonetic fact has made some phonologists object to the classification of /h/ as
[+Low] ( c f . Fant 197 3 : 181 ) .
98

'-r�hhonetic
is meanslevelthatonlthey whespecifn thiecationoccofurs befasore+Low] isBefore
/h/
/h/ [
/a / .
true onotherthe vowels
thetive becomessegmen[t-Lobecow , m+Hiesgh[-]Low].beforeMorehighsvowels
/h/ pecificandally, -theLow,gl-Hiottalgh] before
[
frica­
LoHiwgh LowHigh fClLOWHigh]
mid vowels. This universal) phenomenon can be formulated as follows:
(

:?- ru l e : P lace ad j u s tment of / h /

(8 ) f l i- l
l:- J -----+
l£x J
+Cont
-Syll +Syll
-Cons
The rul e s a ys t
/h/h at [ ( -Co n -Cons
sonant, -Syllabic, +C o nti n uant, +Low,
-fHoillgh])owinbecog vowelmes mus[-Lotwalso] befoagreere non-lon othew vowels.
feature hiThegh. segment and the
/h/

5. 4 . NATURAL I SAT I ON RULES

In subsection
as:Lntroduced
underlyi ningtosegmethents,soundviz.wesystsaewmfoofurBInon-through
2.1.2. ,
Iff, /z/, /s/
natiandve BI cowhichnsonantshaveposited
lxi , been
ds to beThisubss tdoesitutednotwithmeansotmhate othertherelsoundoisanwords.
lnati,oundsve BItenitems. which Eacoccuhrsofithese
one-tuotes.-one corres­ n
beenl'eprespoientntedtwooutor morein subsec' foreigtionn ' sounds,thator athesinothergle ' nwayative'round,soundIt onehasmay
pondenc e between the l o an s o unds
2.1.2.
and the nati v e s u bstit
'thisforeigreasn ' on,soundit beiwillngberepresmuchentcleearerd by ttowoaccount
or morefor' natitheve'phonolsounds.ogicalFor
processes
section involvseinparately.
2.1.2.
g each of the non-native consonants posited in sub­
5 . 4 . 1. NATUR A L I S A T I 0N O F I fI

a
The phonological
voiceless bilabi a l costopnsonant inismostusuallyloanwords.realiseManyd phonloetican aitellmysas
( [p ] )
/f/

contai n i n g /f/t h e co n sonant have b e n well establis h ed t o t h e extent

withthatitemstheycitandednoinspelt
[p]
longerwithsoundtheorletter
( 5 . 14 : 1 ) and
look foreigsucnh, asi. e . they arebelopronou
p,
( 5 . 14 : 2 ) are ( 5 . 14 )
borrowed f r om w.
Arabic
nced that
Notice
and Dutch
( or English ) respectively.
( 5 . 1 4 ) 1 . p i k i r / f i k i r / [ p i k L r . f i k L r ] ' to t h i n k '
p a h a m / f a h a m / [ p a h a m . f a h a m ] ' to compre h e n d '
 99

p a s a l I f a s a l l [ p a s a l , f a s a l ] ' c hap t e r '

h a p a l I h a f a l l [ h a p a l , h a f a l ] ' to memori s e '
s i pa t I s i f a t l [ s i pa t , s i fa t ] ' c harac t e r '
taps i r Itafs i rl [ ta ps � r , tafs � r ] ' t o i n t erpre t '
s a r a p , s a r a f I s a r a f l [ s a r a p , s a r a f ] 'nerve s '
tarap , taraf Itarafl [ ta rap , tara f ] ' s tage '
a r i p , a r i f l a r i f l [ a r � p , a r � f ] 'wise, in t e l ligen t '
2. p a b r i k I f a b r i k l [ p a b r L ? , f a b r L k ] ' manufac ture '
p a k a n s i I f a k a n s i I [ p a k a n s i , f a k a n s i ] ' vaca t i o n '
s o p i r I s o f i r l [ s o p � r , s o f � r ] ' dr i v e r '
d i p a n I d i f a n l [ d i p a n , d i f a n ] ' di van '
tar i p Ita r i fl [ ta r � p , t a r � f ] ' tariff '
a k t i p l a k t i f l [ a k t L p , a k t � f ] 'active '
p a s i p I p a s i f l [ pa s � p , p a s � f ] 'passive '
i n tens i p l i n tens i fl [ i n tens � p , i n t e n s � f ] ' in t e n s i v e '
mo t i p , mo t i f I m o t i f l [ mo t � p , m o t � f ] ' m o t i f '
po t o , f o t o I fo t o l [ po t o , f o t o ] 'photo '

Despite the strong t e ndenc y to s u bstitute for

[p] i n
[f] loanwords,
asulthearyunderlitemson ythinewhichgphonareeticin exampl
[f]
IfI
level.es This isbelobecwausise morethey arecommospecial
( 5 . 15 )

useedlanguages.
mainly by educItemastedof people
nly realised
voc a
whoarehappenborrowed,
to be b­
fmostlamiliyarfrwiomtWestern h the sourclanguages (
this
i. e . English or Dutch . type )

( 5 . 1 5 ) v a k I f a k l [ f a k ] ' s c h o o l subj e c t '

v i t a m i n I f i t a m i n l [ f i t a m � n ] ' v i tamin '
v e t o I f e t o l [ f e t o ] ' v e to '
v i s a I f i s a l [ f i s a ] ' v i sa '
f a k t a I f a k t a l [ f a k t a ] ' fa c t s '
f a k t o r I f a k t o r l [ f a k t � r ] ' fa c t o r '
f a n a t i k I f a n a t i k l [ f a n a t � k ] ' fan a t i c '
f e d e r a s i I f e d a r a s i I [ f e d a r a s i ] ' fe dera t i on '
v o k a l I f o k a l l [ f o k a l ] ' vo c a l sounds ( v o we l s ) '
v a r i a s i I f a r i a s i l [ f a r i a s i ] ' varia t i on '

Thspirantis
e naturalisatio n of
I fI discuss e d above ma y b e referred to a s a de­
applin they toleaitexictionmon.s process.
cited in Since thethosedespirantis
( 5 . 15 ) items mustationbe marked
process' -doesP-rulnote 9'
10 0

P-ru l e : Despiranti sation ( Natural i sation)

(9) +Continuant -Continuant +Cons

[ J ----+ [ -----­ optional
J /
of / f / ( )

+Ant-Cor
The rul e says that a voiceless labial f r -
icV oice
a tive +C o nti n uan t, ([
+Consonant,
stop -Conti+Anteri
([ nuant,or,+Cons-Coronal, -Voiceor,ma-Coronal,
onant, +Anteri y become a-Vvoiceless
oice . labial
J)
J)

5 .4.2. NATURA L I S AT I O N O F /z/

The underl y i n g consonant /z / may be phoneticall

less lamino-alveolaInr frisomecatiloanwove rds, ortheaunderl y realis e d as a voice­
voicedyinalgveolar-ispmoalaretalcommonly
([sJ)
affricate
realised ([J J ) .
as edsuasch as such as and in somThise others,phenomenonishasmorbeene
[sJ ( 5 . 16 : 1 ) ,
/z/
/z/
coreflected
mmonly realis [JJ
to some degree in the current writing system . ( 5 . 16 : 2 ) .

( 5 . 16 ) 1 . a z a s , a s a s l a z a s l [ a z a s , a s a s ] ' p r i n c i p l e '
j a z a t , j a s a t / J a z a t l [ J a z a t , J a s a t ] ' b o dy '
i j azah , i j a sah l i J a zahl [ i Jazah , i Ja sah J ' di p l oma '
lazim, l a s i m l 1 a z i ml [ l a z tm , l a s t m ] ' c ommon '
j e n a z a h , j e n a s a h I J a n a z a h l [ J a n a z a h , J a n a s a h ] ' corp s e '
2 . z a m a n , j a ma n I z a m a n l [ z a m a n , J a m a n ] ' e ra '
z a b u r , j a b u r I z a b u r l [ za bo r , J a bo r ] 'psalm '
z a m r u d , j a m r u d I z a m r u t l [ z a m r o t , J a m r o t ] ' emera l d '
z i n a h , j i n a h I z i n a h l [ z i n a h , J i n a h ] ' adu l tery '
z a k a t , j a k a t I z a k a t l [ z a k a t , J a k a t ] ' m o n e y p a i d to m o s q u e '

( butIt lmuses tabove.cobmemonlnoteToyd hoasinwdievercateinthatwhichexamtheplesofunderl

( 5 . 16 : 2 )
) [jJ ( 5 . 16 : 1 )
ying andz/ maasy alsoinbeexamplrealisesed
/
[sJ
theonetwcano 'marknative'in thesoundslexiconis more
co' +mP-rulmonley usedlOa ' aswhena substitute
[sJ is a mo for
r e [zJ,
co m mo n s u bsti t ute for
[zJ a s i n
astheinphoneticand l' e+velabove.
( 5 . 16 : 1 )
( 5 . 16 : 2 )
P-rule lOTheb ' wherulen thatisconverts
[jJ a more commoinntosub stituteor foron
/z/
set up below has to be regarded as an optional rule . [sJ [j J
[zJ
 101

P-rule : Natura l i sation o f

( l O) a . [ +VoiceJ [ -Voic eJ / +Cons

- - - --�
/z/ ( o ptional)
-Son+Cont
+Ant+Cor
b. [+CorAnt [-CorAnt /
-Back
+ContJ -ContJ +Cons
----- �

-Son+Voice
The rule+Vsaoicysethat, +Contithenunderl -
y+Ainngterior, B ac k
consonant+Coronal,
/z/ ([ +-BacConskoJ)nant,may phonet­
-Sonorant,
ically be realis e d as [sJ uant,
( [ + Cons o nant, -Sonorant, - V oic e , +Conti n uant,
+Anterior,
-Continuant,+Coronal, -Anterior,-BackJ)-Coronal,or [}J-Bac( [ k+ConsJ). onant, -Sonorant, +Voice,
5 . 4 . 3.

The NATURA L I SATI ON O F

underlyi n g consonant /5/ /5 / is us u all y realised as a voiceless

ofwrilamiBI.tnino-alg Thissystem,
veolasrtrongfricatetivendency hasonaltheso beenphoneticreflecletveled ibyn themostcurrentspeakers
such as
( [sJ)

( 5 . 17 ) .

( 5 . 1 7 ) s a h , s y a h / s a h / [ s a h , s a h ] ' l. e ga l. '
s a r a t , s y a r a t / s a r a t / [ s a r a t , s a r a t ] ' r e q u i remen t s '
a s i k , a s y i k / a s i k / [ a s L 7 , a S L 7 ] ' to be a b s o r b e d i n '
i nsaf , i nsap , i n sya f / i nsaf/ [ i n sap , i nsaf , i p s a f ] ' to r e a l. i s e '
m a s a r a k a t , m a s y a r a k a t / ma s a r a k a t / [ m a s a r a k a t , m a s a r a k a t ] ' s oc i e ty '

illcanItemubstrated
s cited above
by the areaboveArabicexamplesborrowedmay words.
e formalised as follows: be terme The' fnatrontiuralisati
d n g' oenssofand/s/
proc
102

P-rule : Fronting

r- l
( ll ) AntCo AntCo / ( Natur a l i sa tion )

- - - --� r+ l
li of /s / ( o ptional)
l- rJ rJ
-+CVoiceont
-Back
+Cons
+CasoThenti[s]nruluant,([+eConsonant,
sa-Bacys kthat, -Anteri
the underl
o r, ying /s/ ( ma[ +yConsphoonant,
-Coronal]) n etic
-Voice, +Continuant, -Back, +Anterior, +Coronal]). a -llVoice,y be realised
5.4.4. NATUR A L I SAT I O N O F Ixl

levelTheassucunderlyi
([k])
voiceless ng consonant
gl o ttal f
h as y realisInedsoasmeacavoiceless
( 5 . 18 : 1 ) .
/x/ricatimayvealso( [ h])beorrealised
a on thevelphonetic
voiceless
se s, the undeglottalrlyinfricg consonant a r stop
/x/or
mayvoicedbe phoneticall
a velar stop ( [ g]) as in ( 5 . 18 : 2 ) .
a tive ( [h ] )

( 5 . 1 8 ) 1 . k h a b a r / x a b a r / [ x a b a r , h a b a r , k a b a r ] ' ne w s '
a k h i r / a x i r / [ a x L r , a h L r , a k L r ] 'end '
k h u s u s / x u s u s / [ x u s o s , h u s o s , k u s o s ] ' sp e c i a l '
k h i l a f / x i l a f ! [ x i l a f , h i l a p , k i l a p ] ' to be wrong '
k h a s / x a s / [ x a s , h a s , k a s ] ' ty p i ca l '
i kh l a s / i x l a s/ [ i x l a s , i h l as , i k l a s ] ' s i n c e re '
m a k h l u k / m a x l u k / [ ma x l o ? , m a h l o ? , m a k l o ? ] ' human b e i ng '
t e k n i k , t e h n i k / t e x n i k / [ t e x n L k , t e h n L k , t e k n L k ] ' te c h n i q u e '
2 . b i o l o g i / b i o l o x i / [ b i o l o x i , b i o l o h i , b i o l o g i ] ' b i o l o gy '
s o s i o l o g i / s o s i o l o x i / [ sos i o l ox i , sos i o l oh i , sos i o l og i ] 'sosiology '
g eog r a f i / xeog r a f i / [ xeog ra f i , heog r a p i , geog r a p i ]
w a g o n , w a h o n / wa x o n / [ wa x � n , wa h � n , w a g � n ]
r e g l em e n / r e x l e m e n / [ r e x l e m £ n , r e h l e m £ n , r e g l e m £ n ]

Iwhichtems ciisteda Dutch/English

in arborrowed
( 5 . 18 : 1 ) e Arabicword),borrowedandwordsthose (ecitedxceptinthe last item( 5 . 18 : 2 )
areThDutce aboveh/Englishexamplesborrowedshow words.
that /x / ma y phonetically b e realised a s [hJ
in allcacassese s,( corf. as in Isonmote hcaerseswords,( c f. the ruleandconverti
other [kJ
5 . 18 : 2 ) .
5 . 18 : 1 ) as ngithen some
[gJ

uapplnderlyinyigngtocoinsteomnants cited/x/ininto ( orand[hJ)themuslitke,beandpreventviceedversfroam, the

[kJ
( 5 . 18 : 2 )
rulebe prevented
which converts
from appltheyinunderl g to yiteinmgsconscitedonantin /x/ into and the( or like. musTot
( 5 . 18 : 1 )
[gJ [hJ)
 103
solvedthis' +P-rulproblem,
marke e 12a' iandtemsthosecitedcitedin ( 5in. 18:( 15 ). 1and8: 2 ) theandltheike havelike haveto beto
be marked ' + P-rul e 12b' in the lexic o n. Since the underlyi n g cons o nant
belomayw hasalsoto bebe regarded
/x/ realised asas an optional
/x/ on the phonetic level, P-rule 12 set up
( 12 ) [+Contcon, � a. [-CoConsnt �
P-rule : Natura l i sation of /x/ ( o pti o nal) rul e .
----� /

b. [-ContCons � -Syll
-+HiVoicegh -+HiVoicghe
c. [+CoConsnt �
++HiVoicegh +Back -Son
--HiVoicghe
+Conti The rul e says that the underlyi
nuant,realised-Voice,as +Hi(g[h,+Cons-Syllabic, n g consonant /x/
-Sonorant, ( [ +
+BackJ) Cons o nant,
ma+Hiygh,be phonet­
icall
-Syll y [kJ o n ant, -Co n ti n uant, - V oice,
+Hi-Voice,gh,abic,-Syllabic,
-Hi
-Sonorant,-Sonorant,
g h, -Syllabic,
+BackJ) +BorackJ) o([r+Conso(nant,[ -Consonant,
+Bac k
[9J

J). [hJ
Thi s rul e ( P -rul
-Continuant,+Contin+Voice,
e l2c) mus t uant,
apply
before P-rule 8.
5.5 . NASAL A S S I M I LA T I ON , SCHWA I N S E RT I ON A N D CONSONANT DELETION
( COAL E S C E NC E )

BI.are IngoiOTablnnegofetothe2deal( saffixes

ee p. 4 5)is, we saw 23' acphonologi
/m a n /
withto thea numberprefix, t ive voic caverbl formsprefiofxaffixes
e
of (morpholthatogicis a-litansdphonetic '.
phonological) Here iwen
processes
tioneinghbouri related
s in varing morphe /m a n /
ous envimesronments and phonological changes i t c a us alterna­
e s in
weruleswilltobeaccount ormulaAs!siiunoggested
concernedfor withtheoralterthesounds.nfations nshapeof threetheby thekindsheadiprefix
/ma n /
of nphonol
g, hereogical
under­
neigoestwoghbouri
in varingomorphemes
us environments or s o and forForphonolthisopurpos
unds.
kinds of morphological environments in which may occur,
gical echanges
/ma n /
, we it causes
need to exami inne
104

namely (1) man / / bef

' noun forming prefix' and stems ( c f. Section o re ste m s, and
( 2 ) /m a n /
3.4. ).
inter m edi a t e bet w een /p a /
5 . 5 . 1. / m a n / B E F O R E S T E MS

Th
ginninbefore e /man / prefix
g with sstonorant is pho n etic a lly realised a s
[ma ]
consnninonag withnts andlabialsonorantobstrglides bef o re
( cf.f. asas
5 . 20 ) ,
s t e ms be­
=mam]
beforebefsteomsre stems e m s begi
beginnibeging nwinitnhgdentwithal/( alpalveolar u ents
obstruents ( c
5 . 21 ) ,
( c f.
5 . 22 ) ,
[ ma n ]
as
=map ]
palginanitalng withfricavowels, tive convelsonantsar and( cglottal f. aobstruents
5 . 23 )
tal)and affas ricate,beforealveolarstemands be­
[ ma Q ]
( cisf. phonologically
5.24) . In other
words,
conditioned. t h e fi n al
( N otenas a l
that seg m
nas e n
a t
l s i n
i t h e
/m a n /
n ste m fi n alprefix
positio n do not undergo
any modification when followed by suffixes, such as
= m a k a n ka h ] ' ea t ? ' , / t o l o Q + l a h / [ t o l o Q l a h ] ' he lp
/ma k a n + k a h /
(p l e a s e ) ' , / t a n a m+ k a n /
= tanamkan ] 'p lant (it) ' . )

outneedBefore above,to exawemtheinexamiechanges

ne furtthatherocthecuralloin msteorphsms follof theowin/man/
another pheno m e n on which is typic a l g tothe the
prefix
pref
/m a n /
i x, pointed
weprefix,
thecase,:L . e . /manthethe/teprefix
ndencywheamonngit BIcomsbipeakers n es withto moinsertnosyllaabicschwasteimmmediatel
s. I n y afh ater
suc
in exampls/man/
asi3yllabic etemss inprefixBI arebelwillveryow. belIirealised
( 5 . 19 ) tmitedmust beinapointed
s outon thhowever,
[maQ ]

number, and mos

e phoneticthatlevelmono­
t of them are
loan items.
: 5 . 1 9 ) / ma n + p a k / [ ma Q a p a k - m a m a k ] ' to p a c k '
/ m a n + t i k / [ ma Q a t � k - m a n � k ] ' t o type '
/ m a n + s a h + k a n / [ ma Q a s a h k a n - m a p a h k a n ] ' to l e ga li s e '
/ m a n + b e l / [ m a Q a b e l - m a m b e l ] ' to ring a b e l l '
/ m a n +c a t / [ ma Q a c a t - m a p c a t ] ' to pain t '
I m a n + l a k / [ m a f) a l a k - m a l a k ] ' to s e a l '
/ m a n + c a p / [ m a f) a c a p - m a p c a p ] ' to s tamp '
/ m a n + b o m / [ m a f) a b �m - m a m b �m ] ' to b omb '
/ m a n + r e m / [ m a f) a r em - m a r e m ] ' to brak e '
/ m a n + l a s / [ m a f) a l a s - m a l a s] ' to we l d '

areareThnot.iedenticphonetic
aWelse( eif. oefrmsu. rthermoof /manre,whereas
[ m a f) ] ) ,
/ in thethosfierswhich
t set appear
of the above
i n the derivations
sec o nd set
and�_n th/se/fiarerstdelseet.tedThisin themeanssecthatothatnd stheethteofsteschwathem initial
above consvations,
deri onants /pbut/ , not/t/
insertion rule must apply
 105

befaboveorederirulesvations.generating the forms appearing in the second set of the

bibinnNotic
eses wiwietthhthatmobisylnosyllthelabicascbichwaiteitemins.mses.rtionIt Itnevernever
processoccursonly wheoccursn thewhenprefix comco­m­
occurs when the
/m a n /

prefi x c o mbi n es
withsyllabisyl
bic s leabicms combior polysyll
t n e with aother
bic steprefixes.
ms. Nor doesIn theit occur sch w a whendeletimoonno­
rule setbecaupusebelthoew f( aormlso in other
/m a n / MEN i s mo r rulattractive.
e es), the form is used rather than
MEN

P-rul e :

(13) 0
Schwa -Cons ( optional) ++CVC++
---- +
inser tion

/ [MEN]
+Syl-Higlh
-+BacLowk
where MEN =
-
/ma n /
R ound ' bicactivstee mvois ce verb prefix' and
The ++Crul V C
e + +
=

s a ys mo
that n
/a/
os y ll a ( [ - Cons o nant, +Syllabic, -Hi g h, -Low, +Back,
-thRaound])
tlimitmonosyll may beabicinssteertedms tobetwhichween the and prefix
/m a n /
/m a n /
monosyllabicc a n be staddedems. are( Note
edExampltoes those thatbelhaveow illtheustratestructurethat CVC - isseerealised
( 5 . 20 ) /m a n /
Section as on
3.1. )
[ ma ]
tconsonants
he phoneticandlevel( sonwhenorant)itglcoimdes.bines with stems beginning with sonorant
( 5 . 20 ) 1 . / m a n +m i n t a / [ ma m i n t a ] ' to a s k ( fo r ) '
/ ma n +m a k a n / [ m a m a k a n ] ' to e a t '
/ m a n +m u a y / [ m a m u a y ] ' to expand '
/ m a n + n a n t i / [ m a n a n t i ] ' to wai t ( fo r ) '
/ m a n + n i l a y / [ m a n i l a y ] ' to e v a lua te '
/man+pap i / [ mapap i ] ' to s i ng '
/ m a n + p a l a / [ m a p a l a ] ' t o b la z e '
/ m a n + Q a Q a / [ m a Q a Q a ] ' to open one ' s mou t h '
/ m a n + Q a Q g u r / [ m a Q a Q g o r ] ' to do n o t h i n g ( j o b l e s s ) '
2. / m a n + l i h a t / [ m a l i a t ] ' to s e e '
/ m a n + l u k i s / [ m a l u k L s ] ' t o p a i n t ( p i c ture ) '
/ m a n + l a d a k / [ m a l a d a ? ] ' to exp lo de '
/ m a n + r a s a / [ m a r a s a ] ' t o fee l '
/ m a n + r u s a k / [ m a r u s a ? ] ' to de s troy '
/ m a n + r a k a m / [ ma r a k a m ] ' to record '
106

3. I m a n +w a k i l + i / [ m awa k i l i ] ' to repre s e n t '

I m a n +w a r i s + i / [ m awa r i s i ] ' to i n h e r i t '
I m a n +wa r t a + k a n l [ m awa r t a k a n ] ' t o b roadcas t '
I m a n + y a k i n + k a n l [ m a y a k i n k a n ] ' to c o n v i n c e '
I m a n + y a y a s a n + k a n l [ m a y a y a s a n k a n ] ' to turn i n t o an i n s t i t u ti o n '

[mam]
Examples
bef o re ( s5t. ems21) belbegiownnisnhowg withthatlabial obsis phoneticall
/man/
t ruents. y realisthated asstem­
Notice
initial [p] ( d eri v ed f r om /p/ or /f/- s e e 5. 2 1: 2 , 3 a ) i s del e ted
p(hseeonetic5. 21:level.4 ). NoticePrefix-ialsnoitialthat the steis notm-indelitialeted/fon/ inthe( 5phonetic
[p]
o n t h e
lenotvel
jP-ruleletede 9)on. the phonetic level because it is not realised as ( c f. . 2 1: 3 b ) is
[p]

( 5 . 21) 1 . I m a n + b a 1 i I [ m a m b a l i ] ' to buy '

I m a n + b a c a l [ m a m b a c a ] ' t o read '
I m a n + b u a t l [ m a m b u a t ] ' to ma ke /do '
2. I m a n + p a k a y l [ m a m a k a y ] ' to u s e , to wear '
I m a n + p i 1 i h i [ m a m i 1 i h ] ' to c ho o s e , to v o t e '
I m a n + p a k s a l [ m a m a k s a ] ' to force '
3a . I m a n + f i k i r l [ m a m i k L r ] ' to t h i n k '
I m a n + f i h a k l [ m a m i h a ? ] ' to t a k e s i de '
I m a n + f a h a m+ i l [ m a m a h a m i ] ' to comp r e h e n d '
3b . I m a n + f e t o l [ m am f e t o ] ' to v e to '
I m a n + f i t n a h l [ m a m f i t n a h ] ' t o s lande r '
I m a n + f o r m i l + k a n l [ m a m f o r m i l k a n ] ' t o make i t forma l '
4. I m a n + p a r + b a s a r l [ m a m p a r b a s a r ] ' to en large '
I m a n + p a r + t i 1) 9 i I [ m a m p a r t i 1) 9 i ] ' to h e i g h t e n '
I m a n + p a r + p a p J a l) l [ m am p a r p a p J a l) ] ' to leng t hen '

Examples ( 5 . 2 2) below s h o w that

/ma n / is phonetically
ng witnh dental real
stops. i sed as
Notice
[man ]
texampl /t/ stestmems-inbegiitilaenvel.lnipositio
hat thewhenes undeisitdelrcolyemtedingbineson withthine phonetic ( N ote i n the
that foll
/m a n /
o wi
is n g so
al
realis e d
( 5 . 22) 1 . as [ man ] bef o re
/5/ i n l o anwords
I m a n + d a p a t l [ m a n d a p a t ] ' to g e t , t o r e c e i v e '
- s e e examples ( 5 . 2 5).)
I m a n + d i d i k l [ m a n d i d � ? ] ' to e du c a t e '
I m a n + d a k i / [ m a n d a k i ] ' to c li mb (moun t a i n ) '
I m a n + d a l) a r l [ m a n d a l) a r ] ' to h e a r , t o l i s t e n '
I m a n + t u t u p l [ m a n u t Q P ] ' t o c lo s e '
I m a n + t u 1 i 5 1 [ m a n u l L s ] ' to wri t e '
I m a n + t a t a p l [ ma n a t a p ] ' to s e t t l e'
I m a n + t o n t o n l [ ma n o n t o n ] ' to wa t c h '
 107
Examples
[ ma p ] when it( 5co. 2m3)binbelesowiw tshoh wstethmastbegi/mannni/ nisg withphoneticall
( p al a y realisaffricedaaste
tal)
andfromal/5v/eolorar/pal/5/) aistaldelfreicative ted o n consthe phonetic
onants. Noticel e vel. that (derived
[5]

( 5 . 23) la . / m a n + J u a l / [ m a p J u a l ] ' to s e n '

/ m a n + J a m p u t / [ m a p J a m p o t ] ' to fe t c h '
/ m a n + J a g a / [ m a p J a g a ] ' t o t a k e care '
/ m a n + J a w a p / [ m a p J a w a p ] ' t o answe r '
lb . / m a n + c a r i / [ m a p c a r i ] ' to l o o k for '
/ m a n + c u r i / [ m a p c u r i ] ' to s t ea l '
/ m a n + c u l i k / [ ma p c u l i ? ] ' to k i dnap '
/ m a n + c u c i / [ m a p c u c i ] ' to w a s h '
2. / ma n + s a p u / [ ma p a p u ] ' to sweep '
/ m a n + s i k s a / [ m ap i k s a ] ' to t o r ture '
/ m a n + s e r a l) / [ m a p a r a l) ] ' to a t tac k '
/ m a n + s u r u h / [ ma p u r u h ] ' to s e n d/orde r s o m e o n e '
3. / ma n + s a i r / [ m a p a i r ] ' to c o mp o s e a p o e m '
/ m a n + s a r a t + k a n / [ m a p a r a t k a n , m a p s a r a t k a n ] ' to s e t up a s
r e q u i re me n ts '
/ m a n + s a h + k a n / [ ma p a s a h k a n , m a p a h k a n , m a p s a h k a n ] ' t o l e ga l i s e '

Examples
[ ma l) ] bef o re ( 5st. e2m4)s belobegiwnnishongwwiththatvowels,
/man/ isvelarphonetically
and gl o realisobstruents.
ttal ed as
Notice
l(e5vel. 24) of1that. representati( derivedon. from /k/ or /x/) is deleted on the phonetic
[k]

/ m a n + a m b i l / [ m a l) a m b \ l ] ' t o t a k e '
/ m a n + u k u r / [ m a l) u k o r ] ' t o m e a s ure '
/ m a n + i k a t / [ m a l) i k a t ] ' to t i e '
2. / m a n + h i b u r / [ m a l) h i b o r ] ' to en t e r t a i n , to p l e a s e '
/ m a n + h i 1 a 1)/ [ m a l) h i l a l) ] ' to d i s app e a r '
/ m a n + h u k u m / [ m a l) h u kom ] ' t o pun i s h '
3. / m a n + g a n t i / [ m a l) g a n t i ] ' t o change '
/ m a n + g a n g u / [ m a l) g a l) g u ] ' to di s turb '
/ m a n + g i g i t / [ m a l) g i g \ t ] ' to b i t e '
4. / m a n + k a s i h / [ m a l) a s i h ] ' to g i v e '
/ m a n + k a t a + k a n / [ m a l) a t a k a n ] ' to s a y '
/ m a n + k u n i l) / [ ma l) u n \ l) ] ' b e c ome y e l low '
5. / m a n + x a b a r + k a n / [ ma l) a b a r k a n , m a l) h a b a r k a n , m a l) x a b a r k a n ] ' t o
i n form, to s e n d m e s s a g e '
/ m a n + x u s u s + k a n/ [m a l) u s u s k a n , m a l) h u s u s k a n , m a l) x u s u s k a n ] ' to
sp e c i a l i s e , to c o n c e n tra te on a p ar t i c u l a r s ub j e c t '
/ m a n + x a w a t i r + k a n/ � a l) a w a t i r k a n , m a l) h a wa t i r k a n , m a l) x a w a t i r k a n ]
' to worry '
108
The fac t s i l lustrated b y e xamp l e s ( 5 . 2 0 ) t hrough ( 5 . 2 4 ) above c an
be summed up as fo l l ows :

( i ) The final na s a l s e gment in the p r e f i x /m a n / i s d e l e t e d when

t he pre f i x c omb ines with st ems b e ginning w i t h non-s y l la b i c s on ­
orant s e gment s ( i . e . t r i l l , l a t e ra l a n d n a s a l c on s onant s and
vo i c e d g l i d e s - c f . P-rule 1 4a ) .

( i i ) The final n a s a l s e gme n t s in t he p r e f i x /m a n / b e c ome s homorganic

w i t h s t e m- i n i t i a l ob s t ruen t s ( c f . P-rule 1 4b ) , and s t em - i n i t i a l
vowe l s ( c f . P-rule 1 4 c ) . But when the / m a n / p r e f i x o c c u r s
b e fore s t em-in i t i a l / 5 / , t he final n a s a l in / m a n / i s phone t i c ­
a l l y re a l i s e d a s a p a l a t a l n a s a l ( [ p ] ) rather t han a s a n a l v e o l a r
nasal ( [ n ] ) . Thi s i s t rue only w h e n the s t em-in i t i a l / 5 / under­
goe s d e l e t i on ( c oa l e s c e n c e ) .

( i i i ) The final n a s a l s e gment in the p r e f i x /m a n / and a s t em-ini t i a l

vo i c e l e s s s t op c oa l e s c e t o a n a s a l whi c h i s homorgani c w i t h t he
s t op in que st ion ; but when the p r e f i x / m a n / o c c ur s b e fore s t em­
i n i t i a l / 5 / , t he fina l n a s a l in /m a n / and the st em- i n i t i a l / 5 /
b e c ome a p a l a t a l n a s a l ( [ p ] ) .

I t mus t b e pointed out however t hat t here i s a st rong t endenc y among

3peakers of BI to mai n t a in s t em-in i t i a l vo i c e l e s s s t o p s and t he fri c ­
at ive / 5 / o n t h e phone t i c l e v e l i n many l o anwords such as :

( 5 . 2 5 ) / m a n +T d R J d M A H + k a n / [ m a n t a r J am a h k a n ] ' tran s l a t e ' ( Arab i c )

/ m a n + K O O R D I N A S I / [ m a n k o ? o r d i n a s i ) ' to c o - o rdi n a t e ' ( Engl i s h )
/ m a n + P d RA KT E K+ k a n / [ m a m p a r a k t e k a n , m a m p r a k t e k a n ] ' to prac t i s e '
( Engl i s h )
/ m a n + S U P S I D I / [ m a n s u p s i d i ] ' to s u b s i di s e ' ( Engli sh/Dut c h )
/ m a n + S E N S U S / [ m a n s e n s o s ] ' t o condu c t a c e n s u s ' ( Engl i s h )

The d i s s imi lat i on involved in the re a l i sat i on o f t he f i n a l n a s a l i n

o;he pre f i x / m a n / as [ p ] when fused w i t h s t em-in i t i a l / 5 / ( c f . 5 . 23 : 2 )
e an b e e xp l ained as fo l l ows . S i n c e t he fina l n a s a l s e gment in the
pre f i x /m a n / b e fore s t em-in i t i a l nasal s , and s t em-in i t i a l v o i c e le s s
e o n s onan t s f o l l owing / m a n / ( e x c e p t / � / ) are subj e c t t o de l e t i on ( or
e oale s c e nc e ) , t here are qui t e a numb e r o f derivat i on s whi c h are morpho­
l o gi c a l l y amb i guous b e c au s e they are homophoni c whi c h , i n turn , c an
eause addi t i onal prob l ems in l earning s u c h as ( 5 . 2 6 ) b e l ow .

( 5 . 2 6 ) 1 . mema d a t [ mama d a t ] - a. / m a n + PA DAT/ ' t o b e come s o l i d '

b. / m a n +M A D A T / ' to s m o k e o p i um '
m a m a s a k [ m a m a s a ? ] - a . / m a n + P A S A K/ ' to fi x b o l t '
b. /man+MASAK/ ' to c o o k '
 109

2 . m e n o b a t k a n [ manoba t k a n ] - a . /man+TO BAT+ka n / ' t o make s om e o n e

repen t '
b. /ma n + N O BAT+ k a n / ' t o inaugura t e '
m e n a l a m [ m a n a l a m ] - a . / m a n +T A L A M / ' to p u t i n a t ray '
b. /man+NALAM/ ' t o w r i te a p o e m '
3. m e n g u k u r [ m a.l) u k o r ] - a . / m a n + KU KU R /
b. / m a n + U KU R /
' t o g ra te '
' to me a s u r e '
m e n g e r a m [ ma l) a r a m ] - a . / m a n + K d R A M / ' to i mp r i s o n '
b. / m a n + d RA M / ' to ha t c h '
c. / m a n + � d R A M / ' to grumb l e '
4 . m e n y a r i n g [ m a fl a r L I) ] - a . / m a n + S A R I � / ' to s o r t o u t '
b. / m a n +)l A R I � / ' to be come l o ude r '
m e n y a l a n g [ m a fl a l a l) ] - a. / m a n + S A L A tj / ' t o s tab '
b. / m a n +)l A L A � / ' to be wide o p e n ( e y e s ) ,

The learner wi l l c e rt a i n l y have d i f f i c u l t y in knowing whe t h e r t he st ems

of t he above d e r i vat i on s b e gi n w i t h voi c e l e s s ob s t ruents or t he c orre s ­
ponding homorgani c n a s a l s ( or vowe l s i n the c a se o f t he f i n a l n a s a l s e g­
ment in Im a n l i s r e a l i s e d as a ve lar - [ I) J ) . From a l e arning p o i n t o f
view , a derivat i on whi c h invo lve s mor e a l t e rnat ive s ( or amb i gu i t i e s ) i s
more d i f f i c u l t t o learn t han t h e one w i t h l e s s a l t e rnat i ve s . The d i s ­
s im i l at i on whi c h o c c ur s when t he final n a s a l s e gmen t i n Im a n l and s t em­
i n i t i a l l s i c oa l e s c e to a p a l a t a l n a s a l ( [ fl J ) may s e rve t he purp o s e o f
preven t i ng the o c c urren c e o f deriva t i on s w i t h a l t e rn a t i v e s b e tween
stem-init ia l s I n / , I t I and l s i . But t hen , t he l e arner has to fac e
another kind o f morphol og i c a l amb i g u i t y b e c a u s e b o t h s t em-init i a l I n l
and l s i are phone t i c a l l y real i s e d as z e ro when p r e c e d e d b y t he pre f i x
Im a n / . Speakers o f B I minimi s e the l e arn ing prob lems b y preserving
s t em- i n i t i a l l e i on the phonet i c l e v e l when preceded b y Im a n / . Thi s
may e xp la i n t he e xc e p t i onal c harac t e r o f l e i - t he o n l y vo i c e l e s s ( t rue )
c on s onant in nat ive B I whi c h d o e s n o t unde r go c oa l e s c en c e ( or d e l e t i o n )
when p r e c e d e d b y t he Im a n l p r e fi x .
The phone t i c varian t s o f the f i n a l n a s a l s e gment i n Im a n l i l l u s t r a t e d
b y e xamp l e s ( 5 . 2 0 ) t hrough ( 5 . 2 6 ) ab ove c an b e formu l a t e d a s fo l l ow s :
110

(14) [ l--.
P -rule : Na sa l d e letion and a s s im i lation

cons
+Na s a l
+Ant
a. o I M E ----- +
[ jSY '
+Son
+Seg
+Cor
b. aAnt I ME ----- + aAnt
8Cor 8 Cor
yBack yBack
+Nasal -Syll

- [
+Cons -Son
+Seg

J
v . -Ant I t1 E --- - + Ant
- C or -Cor
+Bac k +Syll
+Nasal -Cons
+Cons

whe re ME t he f i r s t two s e gment s o f I m a n l ' ac t ive v o i c e

=

verb p r e f i x ' , S e g s e gmen t ( i . e . c o n s o nant or g l i de )

=

The rule says t hat :

( a ) the final n a s a l s e gment in the p re fi x Ima n l i s d e l e t e d b e fore s t em­

i n i t i a l non- s y l l a b i c sonorant s e gmen t s ( [ -S y l l ab i c , + Sonorant ,
+ S e gment J ) ;

( b ) t he final nasal s e gment in the p r e f i x I m a n l b e c ome s homorga n i c

w i t h stem-in i t i a l ob st ruent s e gment s ( [ -S y l l ab i c , - Sonorant ,
+Segment J ) ;

( c ) t he final n a s a l s e gment in the p r e f i x Im a n l i s phone t i c a l l y r e a l ­

i se d as a ve l a r nasal ( [ +C o n s onant , +Nasal , +Ba c k , -Corona l ,
-An t e r i o r J ) b e fore s t em-i n i t i a l vowe l s ( [ - C o n s onant , + S y l l ab i c ,
- C orona l , -An t e r i o r J ) . Thu s t he . final n a s a l s e gment in Ima n l may
be regarded as homorgan i c with a s t em-in i t ial vowe l to t he de gree
t hat b o t h are [ -Ant e r i o r , - C oronal J .

Not i c e t hat s t e m - i n it i a l vo i c e l e s s c on s onan t s other t han l e i ( and

a l s o loan sounds - [ f J , [5J, [ x J ) in the above examp l e s ( c f . 5 . 21 : 2 ;
5 . 2 2 : 2 ; 5 . 2 3 : 2 ; 5 . 2 4 : 4 ) are d e l e t e d when prec eded b y Im a n / . The d e ­
l e t i on o f stem-init ia l vo i c e l e s s c on s onant s a l s o o c curs i n t he s e c ond
part of re dup l i c at e d forms as i n ( 5 . 2 7 : 1 ) b e low .

( 5 . 27 ) 1 . / m a n + p i l i h + p i l i h / [ m am i l i h m i l i h ] ' t o be s e Z e c ti v e '
/ m a n + p o t o � + p o t o � / [ m amo t � � mo t � � ] ' to c u t rep e a te d Zy '
 III

/ m a n + t a r i + t a r i / [ m a n a r i n a r i ] ' t o dan ce ( b e ca u s e o f exci t e me n t ) ,

/ m a n + t u l i s + t u l i s / [ m a n u l L s n u l L s ] ' to wri te w i t h o u t any def-

i n i te purp o s e '
/ m a n + k a r a Q + k a r a Q / [ m a Q a r a Q Q a r a Q ] ' t o make up '
/ m a n + k a t a + k a t a + i / [ m a Q a t a Q a t a i ] ' to s p e a k i t t a b o u t s o m e o n e '
/ m a n + s a p u + s a p u / [ m a r a p u r a p u ] ' t o s w e e p ( j u s t for t h e s a k e of
sweeping) ,
/ m a n + s i a + s i a + k a n / [ m ar i a r i a k a n ] ' to wa s t e '
2 . / m a n + b a c a + b a c a / [ m a m b a c a b a c a ] ' to r e a d w i t h o u t s e r i o u s a t te n ­
tion '
/ m a n + d a s a k + d a s a k / [ m a n d a s a ? d a s a ? ] ' to k e e p i n s i s t i n g '
/ m a n + g e l e Q + g e l e Q + k a n / [ m a n g e l e Q g e l e Q k a n ] ' to s ha k e o n e ' s h e a d
( r ep e a t e d Z y ) ,
/ m a n + c a r i + c a r i / [ ma r c a r i c a r i ] ' to k e e p s e a r c h i n g '
/ m a n + J a r i t + J a r i t / [ m a r J a r L t J a r L t ] ' t o k e e p s ho u ti n g '
/ m a n + h a l a Q + h a l a Q + i / [ m a Q h a l a l) h a l a Q i ] ' t o k e ep h i nde r i n g '
/ m a n + a l u + a l u + k a n / [ m a Q a l u a l u k a n ] ' to we t c om e wi t h de c o ra ti o n s ,
t o p ra i s e '
/ m a n + l i h a t + l i h a t / [ m a l i a t l i a t ] ' to h a v e a t o o k '

N o t e t ha t s t em-in i t i a l sounds in ( 5 . 2 7 : 2 ) above are n o t a ff e c t e d b y

/ma n / in t he s e c ond part o f t h e redup l i c a t e d forms .
The c oale s c e n c e whi c h o c c urs when / m a n / i s fo l l owed by a s t em-init i a l
voi c e l e s s ob s t ruent i l l u s t rated b y e x amp l e s ( 5 . 2 1 : 2 ; 5 . 2 2 : 2 , 5 . 2 3 : 2 ,

[
5 . 2 4 : 4 , 5 . 2 5 : 1 and 5 . 2 7 : 1 ) above c an b e formu l a t e d a s f o l lows :

P - ru l e : Consonant coa lescence

(15) + ME Ant + a. +Cons ++ ( 5 6 ) ( SUFF I X )

J
x
+Cor -Vo i c e
+ Na s a l - C ont
+Cons aAnt
BCor
-aBack

b. +Cons
-Vo i c e
+Cont
+Ant
+Cor

1 2 3 4 5 6 7 (5 6) (8)

( c ont ' d -+)

112

----+ 1 2 c. aAnt 6 7 ( e. aAnt 6 ) (8)

a C or aCor
-aBac k -aBa ck
+Na s a l +Nasal
+Cons +Cons

d. -Ant f. -Ant
-Cor -Cor
-Ba c k -Back
+Nasal +Nasal
+Cons +Cons

(3 4 5) (3 4 5)
Condi tion : I f a , t hen c and e ; i f b , t hen d and f
ME the f i r s t t wo s e gment s of /m e n / ;
=

X sequence o f s e gme n t s b eginning w i t h

=

a vowe l

The rule says that the final n a s a l s e gment in the p r e f i x / m e n / and

a s t em- i n i t i a l vo i c e l e s s s t o p ( [ +Consonant , -Vo i c e , - Cont inuant ] )
e oa l e s c e t o a n a s a l whi c h i s homorgani c w i t h the s t em-in i t i a l s t op .
Not i c e t hat t he feature s anterior and back whi c h require reve r s e d spe c i ­
f i c a t i o n s b l o c k t he rule from applying t o / e / ( [ +Cons onant , -Vo i c e ,
·-Cont inuant , -Ant erior , -Bac k ] ) . And when the pre fix /me n / o c curs
be fore s t em-ini t i a l / s / ( [ + C on s onant , -Vo i c e , +Continuant , +Ant e r i o r ,
+Coronal ] ) , t he final s e gment in /m e n / and / s / b e c ome s a p a l a t a l n a s a l
: [ p ] - [ +C o n s onant , +Na s a l , -Ant erior , -Corona l , -Bac k ] ) . The rule
s t a t e s fur t h e rmore t hat when the s t e m f o l lowing the /m e n / prefix i s
redup l i c at e d , the s e c ond part o f the redup l i c at e d form a l s o undergoes
t he c oa l e s c e n c e proc e s s . Obviou s l y , the rule must be a l l owed t o a p p l y
t o the o u t p u t o f natura l i s at i on r u l e s s e t u p in S e c t i on 5 . 4 . , so t hat
phone t i c forms i l l u s trated by e x amp l e s ( 5 . 2 1 : 3a , 5 . 2 3 : 3 and 5 . 2 4 : 5 ) may
b e c orre c t l y derive d . Meanwhi l e , l oan i t ems such a s ( 5 . 2 5 ) whi c h do
not undergo c oa l e s c en c e when preceded b y t he /me n / p r e f i x have t o be
narked ' - P-rule 1 5 ' in the l e x i c on .

5. 5 . 2 . / m e n / I N T E R M E D I A T E B E T W E E N / p e / A N D S T EMS

It was argued in Se c t ion 3 . 4 . t hat i t is de s i ra b l e and phonologi c a l ly

p l a u s i b l e t o p o s t ulate the p r e f i x / m e n / in t he unde rlying repre s e n t a t ions
o f words ( de rived nouns ) l ike p e n g e r j a [ p e Q e r J a ] ' ex e a u t o r of a j o b ' ,
p e n y e r t a [ p e p e r t a ] ' a aompany (fri e n d ) ' , p e n d a g a n g [ p e n d a g a Q J ' ( n o n ­
profe s s i o n a Z ) trader ' , p e n d i d i k a n [ p e n d i d i k a n ] ' e duaa t i on ' , p e n i l a i a n
[ p e n i l a y a n ] ' e va Z u a t i o n ' , p e m u a i a n [ p em u a y a n ] ' e xpa n s i o n , e x t e n s i o n ' ,
 113

etc . The ma in c on c ern i n t h i s sub s e c t ion i s t o a c c ount for the phone t i c

a l t ernant s o f t he pre f i x /m a n / in s u c h environmen t s .
With re s pe c t t o st ems , / m a n / p o s t u l a t e d int e rme d i a t e b e tween / p a /
and s t ems i s simi lar t o /ma n / pre fixed t o s t e rn s d i s c u s s e d in sub s e c t ion
5 . 5 . 1 . above in t he s e n s e that the final nasal s e gment in t he /m a n /
pre f i x i s subj e c t t o ( a ) d e l e t ion when o c c ur s b e fo re s on o rant c on s onan t s
and g l i d e s , ( b ) a s s im i l a t i on when o c c u r s be fore st ern-ini t i a l vowe l s and
o b s t ruent s , and ( c ) c o a l e s c e n c e when o c c urs b e fore s t ern- i n i t i a l vo i c e ­
le s s s t o p s and a l ve olar fri c at ive / 5 / . The /m a n / p r e f i x p o s t u l a t e d
b e t w e e n / p a / and s t erns i s d i f ferent from /m a n / a d d e d t o s t erns w i t hout
/ p a / in front in t hat the f i r s t t wo s e gment s of the p o s t u l at e d /m a n /
never appear o n t he phone t i c l e v e l o f repre sent a t i on a s i n e xamp l e s
( 5 . 2 8 ) b e l ow .

( 5 . 2 8 ) 1 . I p a+m a n + b a c a l [ p am b a c a ] ' re a d e r '

I p a +m a n +d i d i k l [ p a n d i d L ? ] ' e duca t o r '
I p a +m a n + J u a l l [ p a r J u a l ] ' t rade r , s e H e r '
I p a+m a n + g a n t i I [ p ,H) g a n t i ] ' s u b s t i t u t e '
I p a +m a n + u J i I [ p a l) u J i ] ' e xami n e r '
I pa+man+c u r i l [ parc u r i ] ' thief '
I p a +m a n + h a r a p + a n l [ p a l) h a r a p a n ] ' e xp e c t a t i on '
2 . I p a+man + p i m p i n l [ p am i m p L n ] ' l eader '
I p a + m a n + t u 1 i 5 1 [ p a n u 1 L 5 ] ' wr i t e r '
I p a+m a n + k a r a l) l [ p a l) a r a l) ] ' co mp o s e r , a u t h o r '
I p a +m a n + s a s a l + a n l [ p a r a s a l a n ] ' re g r e t '
3 . I p a + m a n +m u a y + a n l [ p a m u a y a n ] ' e xpan s i o n , e x t e n s i on '
I p a +m a n + n i 1 a y + a n l [ p a n i 1 a y a n ] ' e va lua t i o n '
I p a +m a n + r a r i I [ p a r a r i ] ' s in g e r ( n o n - p rofe s s i o na l ) '
I p a + m a n +w a r t a l [ p awa r t a ] ' t h e p er s o n w h o i n fo rme d '
I p a+ m a n + l u k i s l [ p a l u k L s ] ' p a i n te r '
I p a+ m a n + r a n t a w l [ p a r a n t a w ] ' trave Z Z e r '

The proc e s s e s involving t he final n a s a l in /ma n / and s t ern-init i a l

sound s , name ly n a s a l a s s imi l a t i on ( c f . 5 . 2 8 : 1 ) , c on s onant c o a le s c en c e
(cf. 5 . 2 8 : 2 ) and n a s a l d e l e t ion ( c f . 5 . 2 8 : 3 ) have b een c ap t ured b y rul e s
s e t u p i n sub s e c t i o n 5 . 5 . 1 . P-ru l e 1 6 s e t u p b e low ac c oun t s f o r the
d e l e t i o n of the f i r s t t wo s e gmen t s o f t he prefix / m a n / interme d i a t e
b e tween / p a / a n d s t erns i l l us t ra t e d b y t he ab ove examp l e s ( 5 . 2 8 ) .
114

P-rule : Deletion o f the f i r s t two segmen ts o f /ma n /

( 16 ) [ME] ----+ 0 / + P E + [ ----] [ 1'

CO"'
+Na s a l
+Ant
STEM ( SUFF IX )

+Cor

where M E the f i r s t t wo s e gment s o f t he p r e f i x /m a n /

PE the p r e f i x / p a /

The rule say s t hat t he f i r s t two s e gment s o f t h e p r e f i x /m a n / are

d e l e t e d when the pre f i x is p o s t ul a t e d b etween /pa/ and s t ems .

5 . 6. VOWE L : LAX I NG , RETRAC T I ON , DELETION , NASAL I SAT I ON ; LOWE R I N G A N D

C OA L E S C E N C E

5 .6 . 1 . V OW E L LAX I NG

In BI vowe l s u s ua l l y b e c ome lax when they o c cur in f i na l c lo s e d

s y l lab le s - i . e . t h e y a r e relat ive ly more o p e n and produced w i t h l e s s
art i c u l a t ory e ffort c ompare d w i t h t h e c orre sponding t en s e vowe l s .
Chomsky and Hal le ( 1 9 6 8 : 3 2 4 ) wri t e :

T e n s e s ou n d s a r e p r o du c e d w i t h a d e l i b e r a t e , a c c u r a t e , m a x ­
i m a l l y d i s t i n c t g e s t u r e t h a t i nv o l v e s c o n s i de r a b l e mu s c ul a r
e f fo r t ; n o n t e n s e s ou n d s a r e pr o d u c e d r a p i d l y and s omewhat
i n d i s t i n c t ly . In t e n s e s o un d s , b o t h v owe l s a n d c o n s o n a n t s ,
t h e p e r i o d dur i n g wh i c h t h e a r t i c u l a t o r y o r g a n s ma i n t a i n t he
a p p r o p r i a t e c o n f i gur a t i o n i s r e l a t i v e l y l o n g , w h i l e i n n o n ­
t e n s e s ou n d s t h e e n t i r e g e s t u r e i s e x e c u t e d i n a s om e w h a t
s u p e r f i c i al m a n n e r .

As t o t he d i f fere n c e in the degree o f openne s s b e t ween a t ense vowe l

and t he c orre sponding lax vowe l , Chomsky and Hal le ( 1 9 6 8 ) wr i t e as
fo l l ows :

I t h a s b e e n ob s e rv e d , f o r i n s t an c e , t h a t t he t o ngue c o n s t r i c ­
t i o n i n t e n s e [ i ] i s narrower than that i n lax [ i ] . Thi s
d i f f e r e n c e i n t o n g u e h e i ght i s s u p e r f i c i a l l y r a t h e r s im i l ar
t o t h a t ob s e r v e d b e t w e e n h i gh [ i ] a n d n o n h i g h [ e ] . The
m e c h a n i sm i n v o l v e d , howev er , i s qu i t e d i f f e r e n t i n t h e t w o
cases . . . . ( p . 32 4 )

The symb o l s [ L J , [ E ] , [ o J and [ o J are used i n t h i s s t udy t o repre ­

sent , re s p e c t ive l y , the lax v e r s i o n s o f t he t e n s e vowe l s [ i J , [e] , [u]
and [ o J . Examp l e s ( 5 . 2 9 ) b e low show t hat t en s e vowe l s b e c ome l a x i n
final c l o s e d s y l l ab l e s .

( 5 . 2 9 ) la . / d i d i k / [ d i d L ? ] ' to e du c a t e '
b. / d i d i k + a n / [ d i d i k a n ] ' t h e r e s u H of e du ca ting '
2a . / p i m p i n / [ p i m p L n ] ' to l e a d '
b. / p i m p i n + a n / [ p i m p i n a n ] ' s up e r i o r , a c t of l eading '
 115
3ab.. I pe n d e k l [ pe n d E ? ] ' short '
I k a + p e n d e k + a n l [ k a p e n d e k a n ] ' a b b r e v i a t i on '
4a.b. I J e J e r l [ J e J E r ] ' to s tandls i t i n a row '
/ J e J e r + k a n l [ J e J e r k a n ] ' t o p Za c e in a row '
5a.b. Idudukl [ dudo? ] ' to s i t '
'
I d u d u k + i / [ d u d u k i ] ' t o s i t on ( s om e t h i ng ) '
6a.b. I m i n u m l [ m i n o m ] ' t o dri n k '
I m i n u m + a n l [ m i n u ma n ] ' dr i n k '
7ab.. I t o n t o n i [ t o n t o n ] ' to wa t c h '
I t o n t o n + a n l [ t o n t o n a n ] ' en te r ta i nm e n t '
8b.a . Ibe l okl [ be l o ? ] ' to turn '
I b e l o k + a n l [ b e l o k a n ] ' b e nd, turn '

Notice
are phoneticall t h at i n t h e
ypresrealised above exa m pl e
athes laxsuvowelss, the l a s t vowels of s t e m s ( a -it
andTheasvoweltenselavoxiwnelsg rulineb-isettemsup e ms)
becaus e of t h e e nc e of
belowandapplieswhichto alarel thalereadysix vowels ffi x
s es.
e t fort h i n Secti o n 2.2.,exc e pt
lal lal

l
P - rule : Vowe l laxing
l a x.
i-
li Jj
( 17 CSYIons ---- [ -Tense ] [ -----] C ++
) .. I

The where
rul e s a C
ys s t a
thatnds vofor
w any
els c
beco o ns
m e o nants
l a x ([ - T ens e ]) when t h e y occur i n
final closed syllables.
5 . 6 . 2. V OW E L R E T R A C T I O N

non-f Theinalfrontsyllables
mid vowelwhelein itisdoesusunotallycarryrealisedstress,phoneticall s u c h as y ( a5s. 30 ). in
[a]

( 5 . 30) lb.a . I b e b a s l [ b e b a s ] ' fr e e '

I k a + b e b a s + a n l [ k a b a b a s a n , k a b e b a s a n ] ' fr e e dom '
2a. I d e s a l [ d e s a ] ' v i Z Zage '
b. I p a + d e s a + a n l [ p a d a s a ? a n , p a d e s a ? a n ] ' v i Z Zage area '
3a . I m a r d e k a l [ m a r d e k a ] ' independe n t '
b. I k a + m a r d e k a + a n l [ k am a r d a k a ? a n , k a m a r d e k a ? a n ] ' in de p e nden c e '
4a. I p e n d e k l [ p e n d E ? ] ' s ho r t '

5a.b.b. I k a + p e n d e k + a n l [ k a p a n d e k a n , k a p e n d e k a n ] ' a b b re v i a t i on '

I b e l a l [ b e l a ] ' to defe n d '
I p a + m a n + b e l a + a n l [ p a m b a l a ? a n , p a m b e l a ? a n ] ' de fe nc e '
116

The unde r l y ing l e i o c c urring i n t he penult ima t e s y ll ab l e o f s t ems

above i s phone t i c a l l y rea l i s e d a s [ e ] in t he a - i t ems ( i . e . in s t r e s s e d
p o s i t ion ) and a s [ a ] o r [ e ] in b - it ems ( i . e . in un s t r e s sed p o s i t ion ) .
I t mu s t b e noted however t hat in rap i d s p e e c h one may hear forms s u c h
as ped e s a a n [ pa d s a ? a n J ( c f . 5 . 3 0 : 2b ) , keme r d e k a a n [ kma r d k a ? a n J ( c f .
5 . 3 0 : 3b ) , kependekan [ kapndekan J ( c f . 5 . 30 : 4b ) , etc . The mot ivat i on
behind t he s e forms i s t he tenden c y t o drop [ a ] rather t han a s t ra i ght ­
forward p ro c e s s - lei d e l e t ion ( c f . sub s e c t io n 5 . 6 . 3 . ) . The s c hwa
re a l i s a t io n of l e i i l l us t ra t e d by examp l e s ( 5 . 3 0 ) above may be t e rmed
' re t ra c t ion ' and can b e formul a t e d a s f o l lows :

P-rul e : Vowe l re traction ( op t i ona l )

2
(18) [ -Back ] ----+ [ +Bac k ] I X C VC v c ( C v c ) ++
o 0 0 0 0
-High
-Low
-Round
+Sy11
-Cons
-Stre s s
where X s t ands for any s e gment sequence or n u l l

The r u l e s a y s t hat un s t re s s ed l e i ( [ -Cons onant , + S y l l ab i c , -Bac k ,

-High , -Low , -Round , - S t re s s ] ) b e fore penult imate s y l lab l e may phon e t ­
i c a l l y b e r e a l i s e d as [ a ] ( [ -Consonant , + S y l l ab i c , -Low , -High , +Ba c k ,
-Roun d ] ) .

5 .6. 3 . S C HWA V E L E T I 0 N

It has b een p o in t e d out in S e c t i o n 3 . 1 . t ha t c on s onant c lu s t e r s in

init i a l p o s i t ion and seque n c e s o f t hree or more c on s onant s in me d i a l
p o s i t i o n a r e prevented on the phono l ogi c a l l e v e l b y insert ing a s c hwa
( / a / ) in the unde r l y ing repre sentat ions of a numb e r of loan it ems ,
re c e n t l y b orrowed e i ther from Eng l i s h or Dut c h , t o b reak up s u c h ' in­
admi s s i b le ' s equen c e s . To derive the c orre c t ( or c ommon ) phone t i c
forms o f t h e s e l oan it ems , s u c h a s ( 5 . 3 1 ) , w e s ha l l n e e d a s c hwa
d e l e t i on rul e .

( 5 . 31 ) 1 . / p a r a k t e k / [ p r a k t £ k , p a r a k t £ k J ' to p rac t i s e , prac t i c e '

/ p a r o k l a m a s i / [ p r o k l a m a s i , p a r o k l a m s i J ' p r o c lama t i o n '
/ k a i i n i k / [ k l i n L k , ka l i n L k J ' c linic '
/ k a l e i m/ [ k l e L m , ka l e Lm J ' c laim '
/ s a t r u k t u r / [ s t r u k t o r J ' s truc ture '
 117

2. l i ndusta r i l ( i ndus t r i . i n d u s t a r i ] ' i n du s try '

l i nsat ruks i l ( i ns t ruks i ] ' i n s truc t i o n '
I i n s a p e k s i I ( i n s p e k s i ) ' in s p e c t i o n '
I k o n s a t r u k s i l ( ko n s t r u k s i ] ' c o n s truc t i o n '
l i n s a t a l a s i l ( i n s t a l a s i ] ' i n s ta l la t i o n '

Not i c e t hat t he p o s t u l a t e d / a / i n some o f t h e above i t ems i s never

real i se d on t he phone t i c l e ve l . Tho s e i t ems must b e marked ' +P-rule
1 9 ' i n the l e x i c on . The s c hwa d e l e t i on rule a l s o a c c ount s for the
phonet i c forms o f long e s t a b l i s hed l oan i t ems ( Sanskrit b o rrowed i tems )
s u c h as ( 5 . 32 ) b elow .

( 5 . 32 ) I p u t a r a l ( p u t a r a . p u t r a ] ' b o y , s on '
I p u t a r i / ( p u t a r i . p u t r i ] ' g i r Z , daugh t e r '
I s a s t a r a l ( s a s t a r a . s a s t r a ] ' Z i tera ture '
Isu taral ( sutara . s u t ra ] 'siZk '
IsaJah taral ( saJahtara . saJaht ra ] ' p r o s p e ro u s '

The main reason for c h o o s i n g the f i r s t s e t o f t he p hon e t i c forms o f

t he ab ove i t ems a s t he i r unde r l y i n g forms i s , apart from t he p o s i t ive
c on d i t i on enumerated i n Se c t i o n 3 . 1 . , t o have a s imp l e r d e s c ri p t i on
b ec au s e we c an then do away w i t h s c hwa i n s ert i on rule . One may wi s h
t o t ake the opp o s i t e p o s i t i o n , that i s h e m a y w i s h t o d o away w i t h
s c hwa d e l e t ion rule . But t h i s approach wi l l c er t a i n l y fai l t o a c c ount
for s c hwa de l e t i on whi c h o c c urs in nat ive B I i t ems i n rap i d s p ee c h ,
s u c h as ( 5 . 3 3 ) below.

( 5 . 3 3 ) I b a l u m l ( b a l om . b l om ] ' n o t y e t '
I t a r a r) 1 ( t a r a r) . t r a r) ] ' b r i g h t , l i gh t '
I s a l a ma t l ( s a l a m a t . s l a m a t ] ' sa fe '
I kapadal ( kapada . kpada ] ' to '
I a n ami ( a n a m . n a m ] ' six '

The s c hwa d e l e t i on pro c e s s i ll u s t ra t e d by e x amp l e s ( 5 . 3 1 ) t hrough

( 5 . 3 3 ) above can be formu l a t e d as f o l lows :

P - ru l e : Schwa d e le tion ( op t i onal )

( 19 ) -Cons
+Sy11
---- � o / X [- --- - J C 2V C ( V C
o 0 0
) ++

-Low
+Ba c k
-Round

where X s t an d s for any sequence o f s e gment s o� nul l ,

C for any c o n s onant and for any vowe l V
ll8
mayoutpTheutbe ruledeletedsaysinthatnon-final([-syllables.
lal Consonant, This+Syllabic,rule also-Low,applies +Back, to-Round])
the
of P-rule 18.
5 .6. 4 . V OW E L N A S A L I S A T I O N

Anot h er phenome n on concerni n g vowels t h at can be obs e rved i n the

follsonant,owih nofgthismanyexamvopleswspeleakerswillshow oftethatndBIwhetoisnbetheanasvowelnasalis
speec
alised. is aprectionededof vowels.
b y a nas Theal con­
( �) . 3 4 ) I m a k a n l [ m a k a n ] ' to e a t '
l e n a k l [ e n a ? ] ' de Z i c i o u s '
l a l) i n l [ a l) i n ] ' w i n d '
I r a w a l [ rawa ] 'souZ '
I m a i n l [ m a i n ] ' t o p Zay '

nasalise abegitionnsiswiofteth anvowelcarriored aoutgliacdre.oss

( 1whe974)n clthaeimfolls thatowingthesyllabl
aThesyllHalidatambla ehowever provi d e n o s u pport to
bothsuch nasvowelsalisaintion does not seem to bearesolelynasalcondi
I ma a f l ' to b e s o rry '
this cl a i m . I t is true
ised bytionedsomebyspeakers t h at buteding
nasal. Exawordsmplestends to bebelownasalised
( 5 . 35 ) show thatby manythe sesquenc t
e irresinpective
-aa-
h e prec
Arabicof
borrowed
the preceding consonant. p eakers
( 5 . 3 5 ) I s a a t l [ s a ? a t ] ' mome n t '
I t a a t l [ t a ? h ] ' ob e di en t '
I ma n f a a t l [ m a n f a ? a t ] ' advan tage , u s e '
I s a f a a t l [ s a f a ? a t ] ' in terce s s i on '
I n a a s l [ n a ? a s ] ' un Z u c k y day '

Notsigne ifthaticancthee andnasatherefore 20

linguistic
lisation ofP-rulvoweels discset upussedbeloabovew hashasto beno regarded
as a low level rule with respec( optional)
P-.ru le : Vowe l nas a l i sation
t to the whole ser:t����of lBI phonology rules.
(20) [ -Nasal] [ +Nasal] Gcons
----+

+Nas����alJ con����J
I X � � y

[+ J [+ J
-Cons
+ L o w -Cons In Arabic words
+Low
where both and stand for any sequence of segments
X Y
 119

The rule says t hat vowe l s ( [ -Con s onant , + S y l l ab i c , -Na s a l ] ) are

( op t i onal ly ) n a s a l i sed when p r e c e d e d b y nasa l s . The vowe l s e q u e n c e
- a a - in Arab i c b orrowed words a l s o t e n d s t o b e nas a l i s ed .

5.6 . 5 . V OW E L L O W E R I NG

I n s ub s e c t i on 5 . 3 . 2 . we s aw t hat there are a numb e r o f doub l e t s

r e s u l t i n g from s y l l a b l e reduc t i on ( i . e . t h e de l e t i o n o f / h / t o gether
w i t h the pre c e ding vowe l ) . There i s another t ype o f doub l e t i n
where c er t a i n vowe l s o c c ur i n var iat ion i n c er t a i n word s .
BI
The t yp e s
o f vowe l s t hat may o c c ur in free variat i o n are ( a ) front vowe l s ( [ i ]
and e e l ) as in ( 5 . 3 6 : 1 ) , ( b ) rounded b a c k vowe l s ( [ u ] and [ 0 ] ) a s i n
( 5 . 3 6 : 2 ) and ( c ) c entral vowe l s ( [ a ] and [ a ] ) a s i n ( 5 . 3 6 : 3 ) .

( 5 . 36 ) 1 . / s a n i n / [ s a n i n - s a n e n ] ' Mo n day '

/ b i a / [ b i a - b e a ] ' tax '
/ l i w a t / [ l i w a t - l e wa t ] ' to p a s s ( by ) '
/ p e n i Q / [ p e n l Q - p e n e Q l ' (an ima l ) l i c e n s e b a t c h '
/ a p o t i k / [ a p o t l k - a p o t e k ] ' c he m i s t '
/ k a n t u Q / [ k a n t o Q - k a n t � Q ] ' p o c ke t '
/ l u b a Q / [ l u b a Q - 1 0 b a Q ] ' ho le '
/ g u a / [ g u a - goa ] 'cave '
/ r u b a h / [ r u b a h - r o b a h ] ' to a l te r '
/pu l i s i / [ pu l i s i - p o l i s i ] ' p o l i ceman '
/ s a r a m / [ s a r a m - s a r a m ] ' fr i g h t e n i n g '
/ p i n t a r / [ p i n t a r - p i n t a r ] ' c lever '
/ t a b a s / [ t a b a s - t a b a s ] ' to c u t off (gra s s , tre e s ) '
/ t a r a s / [ t a ra s - t a r a s ] 'pi t h '
/ t a n t a r a / [ t a n t a r a - t a n t a r a ] ' army '

S i n c e t he free variab i l i t y o f the ab ove p a i r s o f vowe l s i s l im i t e d t o

a numb e r o f i t ems , we need t o mark t he s e i t ems and t he l ike ' +P-rule 2 1 '
i n the l e x i c on , t o indi c a t e t hat P-rule 2 1 s e t up b e low app l i e s t o t hem .

P-rule : Vowe l l owering ( opt i o n a l )

(21)a. [ +H i gh ] ----+ [ -H i gh ] /
-Low
aRound
BBa c k
+Sy l l
In a numb e r of i t ems
-Cons
12 0

b. [ -Low J ----+ [ +Low J I

-High
+Ba c k
-Round
+Sy l l
In a numbe r o f i t ems
-Cons

The rule says t hat ( a ) high vowe l s may b e c ome non-high in a numb e r
o f i t ems , and ( b ) t he c e nt r a l m i d vowe l l a l ( [ -Con sonant , + S y l l ab i c ,
- Low , -High , +Ba c k , -Round J ) may b e c ome l ow ( [ a J - [ -C o n s onant ,
+ S y l l ab i c , +Low , -High , +Ba c k , -Round J ) in a numb e r o f i t ems .

5.6.6. V OW E L C O A L E S C E N C E

There i s s t i l l anot her t y p e o f doub l e t i n B I whi c h r e s u l t s from t h e

c oa l e s c e n c e o f a l o w vowe l ( /a / ) and a high b a c k vowe l ( l u i ) i n t o a
back mid vowel ( [ o J ) in a numb er o f p o l y s y l l a b i c l o anwo rds s u c h as the
f o l l owing :

( 5 . 37 ) I t a u l a n l [ t a u l a n - to l a n J ' fr i ends ' ( Arab i c )

I t a u r a t l [ t a u r a t - t o r a t J 'Mo s e s ' Law ' ( Arab i c )
I s a n t a u s a l [ s a n t a u s a - s a n t o s a ] ' s afe , p e a c e fu l ' ( Sanskri t )
l a Q g a u t a l [ a Q g a u t a - a Q g o t a ] ' memb e r ' ( Sanskr i t )
I t a u k e l [ t a u k e - t o k e ] ' emp loy e r ' ( Ch i n e s e )
I taugel [ t auge - toge ] ' b e an sprou t ' ( Chine se )
l a u t o l [ a u t o - o t o ] ' mo tor c a r ' ( Dut c h )

S in c e vowe l c oa l e s c en c e may o c c ur only i n a sma l l numb er o f i t ems ,

words c i t e d in ( 5 . 3 7 ) ab ove and t he like mu s t be marked ' +P-rule 2 2 '

[]
in the l e x i c on .

P·-rule : Vowe l coalescence ( opt iona l )

[:�::J
( :� 2 ) ++ X LOW C V ( C ) ++ ----.
+Ba c k +
+Syll +Syll
-Cons -Cons
I 2 3 4

I 2 -High 5 6 7 8
-Low
+Round
+Sy l l
In a numb e r o f
-Cons
loanwords
(3 4)

where C s t ands for any c on sonant and V for any vowe l ;

X may b e n u l l or non-nu l l
 121

The rule says t hat l a l ( [ - C o n s onant , + S y l l ab i c , +Low , +Back J ) and

lui ( [ -Cons onant , +Sy l l ab i c , +High , +Back J ) may c o a l e s c e into [ o J
,

( [ - C on s onant , + S y l l ab i c , -High , -Low , +Round J ) in a numb e r o f p o l y ­

s y l lab i c l o anword s .

5.7. GL I DE I N S E RT I ON ANO D E S Y LLAB I CAT I ON

5 . 7. 1 . G L I VE I NS E R T I ON

We s a w in sub s e c t ion 5 . 2 . 1 . t ha t a g l ot t a l s t op i s i n s e r t e d t o b r e a k
u p seque n c e s o f two ident i c a l vowe l s on t he phon e t i c leve l . In add i t i on
t o that , speakers o f BI a l s o t e nd t o insert a glide t o break up a s e ­
quence o f a h i gh vowe l f o l l owed b y another vowe l . Thi s i n s e r t e d g l i de
always agre e s on t he feat ure roundne s s w i t h t he f i r s t ( hi gh ) vow e l o f
t he s e quen c e i n que s t i o n . Thus a vow e l sequence b e ginning w i t h ! i l
w i l l t e nd t o be b roken up on t he phone t i c l eve l b y i n s e r t i n g the c orre s ­
ponding ( vo i c e d ) non-round g l i de [ y J a s i n ( 5 . 3 8 : 1 ) . And a vowe l s e ­
q u e n c e b e ginning w i t h l u i w i l l tend t o b e b roken u p on t he phone t i c
level b y i n s e r t i n g t h e c o rresponding ( vo i c e d ) round g l i d e [ w J a s i n
( 5 . 38 : 2 ) . Re c a l l t h e d i s c u s s i on i n sub se c t ion 4 . J . 2 . t hat vowe l s e ­
que n c e s e n d i n g in a m i d vowe l are very rare . Mo s t o f t he e xamp l e s given
be low c ontain s e quenc e s ending in a low o r h i gh vowe l .

( 5 . 38 ) l . / s i ap/ [ 5 i a p - s i y a p ] ' re a dy '

/ t i a p / [ t i a p - t i y a p ] ' every, each '
/ C i u rn / [ c i o m -
c i y orn ] ' t o k i s 8 , t o sme L l. '
/ t i u p / [ t i o p - t i y o p ] ' to b L ow '
/ b i o l a / [ b i 0 1 a - b i yo 1 a ] ' v i o L i n '
/ h i e r a k i / [ h i e r a k i - h i y e r a k i ] ' hi erarchy '
2. / b u a t / [ b u a t - b uw a t ] ' t o d o , to ma k e '
/ d u a / [ d u a - d u w a ] ' two '
/ u a Q / [ u a Q - u w a Q ] ' mo n e y '
/ k u i l / [ k u L l - k u w L l ] ' te mp L e '
/ J u i t a / [ J u i t a - J u w i t a ] ' sw e e t h e a r t '
/ k u e / [ k u e - k u w e ] ' ca k e '

The glide i n s e rt i on proc e s s i ll u s t rat e d by e x amp l e s ( 5 . 3 8 ) ab ove c an

be forma l i se d as f o l l ows :
122

P-rul e : Glide insertion ( op t i ona l )

[ J [+� ��] r :� �� ]
(23) o - - - - +- Cons / X -- - y
-Sy l l
+Vo i c e +High < +High>
aRound aRound --
< -aRound>

whe re b o t h X and Y may b e nul l and non-nu l l

The rule says t hat a glide may opt i ona l l y be i n s e r t e d be tween a

high vowe l ( [ -Con sonant , +Syl lab i c , +High ] ) and another vowe l , and t he
i n s e r t e d g l i de mus t agree on the feature roundn e s s w i t h t he f i r s t ( high )
vowe l . The feature round for the f i r s t and the s e c ond vowe l s mu s t have
opp o s i t e va lue s if the s e c ond vowe l has the feature s p e c i fi ca t i on
[ +High ] .

5 .7. 2 . V E S Y L L A B I C AT I ON

In rapid spee c h , high vowe l s are usua l l y sub s t i t ut e d by the c orre s ­

p Jnding v o i c e d g l i d e s when fol lowed b y o t h e r vowe l s e sp e c i a l ly in p o l y ­
s y l lab i c words in un s t re s s ed p o s i t i on a s in ( 5 . 3 9 ) , and i n a numb e r o f
words b e ginning w i t h high vowe l s f o l lowed b y o t h e r vowe l s as in ( 5 . 4 0 ) .

( 5 . 3 9 ) 1 . / b i a y a / [ b i a y a - b y a y a ] ' exp e n s e s '

/ s i a p a / [ s i a p a - s y a p a ] ' wh o '
/ p i u t a l) / [ p i u t a l) - p y u t a l) ] ' c re di t, loan '
/ s i u m a n / [ s i u m a n - s y u m a n ] ' to r e c o v e r from uncon s c i o u s s ta t e '
/ s o s i o l ox i / [ s o s i o l og i - s o s y o l o g i ] ' s o c i o logy '
/ h i e r a k i / [ h i e r a k i - h y e r a k i ] ' hi e rarchy '
/ s u a t u / [ s u a t u - s w a t u ] ' on e , s om e t hi ng '
/ s u a s t a / [ s u a s t a - s wa s t a ] ' p r i v a t e ( i n s t i t u t i on ) '
/ k a l u a r g a / [ k a l u a r g a - k a l wa r g a ] ' fami l y , re l a ti v e '
/ k u i t a n s i / [ k u i t a n s i - kw i t a n s i ] ' r e c e i p t '
/ k u a l i t e t / [ k u a l i t e: t - kwa l i t e: t ] ' q ua li ty '
/ k u i n t a l / [ k u i n t a l - kw i n t a l ] ' 1 0 0 Kg (un i t ) '

( �) . 4 0 ) / u a l) / [ u a I) - w a l) ] ' m o n e y '
/ u a k / [ u a ? - wa ? ] ' unc l e , aun t '
/ u a p / [ u a p - wa p ] ' s te a m '
/ i a / [ i a - y a J ' to s a y y e s '
/ i u r a n / [ i u r a n - y u r a n ] ' c on tri bu t i on '

Examp l e s ( 5 . 3 9 ) sugge s t t hat ' de s y l l a b i c at ion ' ( i . e . a proc e s s

whe reby a s y l l ab i c segment ( vowe l ) b e c ome s non - s y l lab i c on the phone t i c
leve l ) i s c on d i t i oned b y t he ab s e n c e o f s t re s s o n t h o s e h i gh vowe l s .
 123

Th i s imp l i e s t hat s t re s s p lac ement r u l e s mu s t apply b e fore t he d e s y l ­

lab i c a t i on rule . Examp l e s ( 5 . 4 0 ) however do n o t sat i s fy s u c h c ondi t ion .
Nor do e xamp l e s ( 5 . 4 1 ) b e low .

( 5 . 4 1 ) Isu r i a l [ su r i a - su rya ] ' t he s u n '

I d u n i a l [ d u n i a - d u n y a ] 'wo r l d '
Im u l i a l [ mu l i a - mu l y a ] ' g lorious '
I k a r u n i a l [ k a r u n i a - k a r u n y a ] ' gi ft (from God) '
I k a r i a l [ k a r i a - ka r y a ] 'work '
I m a n u s i a l [ m a n u s i a - ma n u s y a ] ' human b e ing '

Not i c e that stress in the above p a i r s o f a l t e rn a n t s fal l s on t he

s e c ond l a s t vowe l . Thi s c le a r l y s ugge s t s that t he s t re s s p lac ement
rul e s must apply to the output of t he de s y l l ab i c at i on rule set up b e low .
Some o f the p a i r s o f a l t e rnan t s c i t e d above have b e e n re f l e c t e d i n
the wri t ing s y s t e m i n the s e n s e that s ome o f t he ab ove i t ems have t w o
equa l ly - a c c e p t a b l e o rthograph i c forms s u c h a s u a n g / w a n g 'money ' , i a/ya
' to s a y y e s ' , u a k /w a k ' un c l e , aun t ' , i u ran/yuran ' c o n tr i b u t i o n (mon e y ) ' ,
sur i a/surya ' t h e s un ' , k u i t a n s i / kw i t a n s i ' r e c e ip t ' , k u a l i t e t / kwa l i t e t
' q ua l i ty ' , e t c . For t h i s r e a s o n , t he phe nomenon i l lu s t ra t e d b y e xamp l e s
( 5 . 3 9 ) t hrough ( 5 . 4 1 ) above i s viewed a s a d e s y l lab i c a t i on proc e s s
rather t han g l i d e i n s ert ion and vowe l d e l e t i o n p ro c e s s e s . The d e s y l l a ­

[----J l
b i c at i on phenomenon may b e formu l a t e d a s f o l lows :

P-rule : Desy l l abication ( op t iona l )

(24 ) [ +S yl l J [ -S y l l ] / j }
<C> -cons

J
a
----+

CVC -Cons +Syll

++ +High < +H i gh >
c
aRound < -aRound>

Conditions : I f a then b ; i f c t hen d

C s t an d s for any c on s onant , V for any vowe l
and X may be nul l or n on-nu l l

The ru l e says that a h i gh vowe l may opt i on a l l y b e c ome a glide ( non­

s y l l ab i c ) when f o l lowed b y another vowe l i n p o l y s y l lab i c w or d s or in
( po l y s y l lab i c and b i s y l l ab i c ) words b e gi n n i n g with a h i gh vowe l . The
feature round for the f i r s t and s e c ond vowe l s must have reve r s e d s pe c i ­
f i c a t i o n s i f the s e c on d vowe l i s [ +H i gh J .

5.8. VOWEL D I S S I M I LAT I ON

It has b e e n s t at e d in C hap t e r 3 that in B I a word c an be c on s t ru c t e d

b y me an s o f redup l i c at i on o f a s t e m . B y d e f i ni t i on , t he und e r l y i n g
f o r m o f a w o r d c on s t ru c t e d b y redup l i c at i on c on s i s t s o f t wo same p a rt s .
124

In most c a s e s , b o t h p a r t s o f a redup l i c a t e d form a r e a l s o ident i c al on

the phone t i c leve l . There are however a numbe r o f words c on s t ru c t e d
by means o f redup l i c a t i on o f s t ems who se phone t i c forms may n o t c on s i st
o f two i dent i c a l part s e a c h b e c au s e one o f t hem may undergo a vowe l
c hange . Examp l e s ( 5 . 4 2 ) b e low show t hat when t he last s y l l ab l e o f a
s t e m c on t a i n s a h i gh vowe l ( c f . 5 . 4 2 : 1 -5 ) or mid vowe l ( c f . 5 . 4 2 : 6 )
then t h i s vow e l w i l l b e phone t i c a l l y rea l i se d as a low vowe l ( [ a J ) i n
the f i r s t p a r t o f a redup l i c a t i on ; meanwh i l e when t he s e c ond l a st s y l ­
lab l e c on t a i n s a low vowe l ( /a / ) , t h i s vowe l wi l l b e phone t i c al l y
real i s e d as a mid b a c k vowe l ( [ o J ) in the f i r s t part o f t he re dup l i c at i on
(cf. 5 . 42 : 1 -3 ) .

( 5 . 4 2 ) 1 . / b a l i k + b a l i k / [ b o l a ? b a l i ? ] ' to and fro '

2 . / g a n t i + g a n t i / [ g o n t a g a n t i J ' c o n t i n u ou s ly chang i n g '
3 . / p a n t i Q + p a n t i Q / [ p o n t a Q p a n t i Q ] ' t o run as fa s t as o n e c ou l d '
4 . / s a l i Q + s a l i Q / [ s a l a Q s a l i Q ] ' to a l t e rn a t e '
5 . / d a s u s +d a s u s / [ d a s a s d a s o s ] ' rumour '
6 . / c o r e t + c o r e t / [ c o r a t c o r e t ] ' s cra t c he s '

Examp l e s ( 5 . 4 3 ) b e l ow show t hat when t he l a s t s y l la b l e o f a stem

c on t a i n s a l ow vowe l ( /a / ) t hen this vowe l wi l l b e phone t i c a l ly r e a l i s e d
as a front h i gh vowe l ( [ i J ) i n t he s e c ond p a r t o f a redup l i c a t i on ; mean­
whi l e when t he f i r s t s y l l a b l e of t he s t em c o n t a i n s a round b a c k vowe l
( l u i - c f . 5 . 4 3 : 1 or I i i - cf. 5 . 4 3 : 2 , 3 ) , t h i s vow e l w i l l be phonet i c ­
a l ly r e a l i s e d as a l ow vowe l ( [ a J ) i n t h e s e c ond part o f t he derivat ion .

( 5 . 4 3 ) 1 . / 1 u n t a Q + l u n t a Q / [ l u n t a Q l a n t i Q ] ' bobbing '

2 . / c o m p a Q + c om p a Q / [ c o m p a Q c a m p i !,) ] ' t orn ( i n rag s ) '
3 . / p o r a k+ p o r a k / [ p o r a ? p a r i ? ] ' di s ordere d '
4 . / g a r a k + g a r a k / [ g a r a ? g a r i ? ] ' m o v e me n t s , b e hav i o ur '
5 . / p am u d a + p a m u d a / [ p a m u d a p a m u d i ] ' y o u t hs '
6 . / b a s a + b a s a / [ b a s a b a s i J ' good manne r s '

Examp l e s ( 5 . 4 2 ) and ( 5 . 4 3 ) show t ha t the l a s t vow e l o f the f i r s t

p art o f a derivat i o n ( redup l ic a t i on ) i s a lways a low vowe l ( [ a J ) .
A c c ordingl y , when the l a s t vow e l o f a s t e m i s not l a / , t he vowe l c hange ( s )
w i l l t ake p l a c e in t he f i r s t part o f a redup l i c a t i on . In such a c a s e ,
t he l a s t vowe l o f the s t em wi l l b e c ome low and t he f i r s t vowe l , i f la / ,
w i l l b e c ome [ o J ( c f . 5 . 4 2 : 1 - 3 ) . And when t he last vow e l o f a s t e m i s
/ a / , t he vowe l c hange ( s ) wi l l o c c ur i n t he s e c ond part o f a deriva t i o n .
I n s u c h a c a s e , the l a s t vowe l w i l l b e c ome [ i J (cf. 5 . 4 3 ) , and t he f i r s t
v Jwe l , i f l u i or 10 / , w i l l b e c ome r a J (cf. 5 . 4 3 : 1-3 ) .
 125

The vowe l c hange phenomenon i l lust rat ed b y e x amp l e s ( 5 . 4 2 ) and

( 5 . 4 3 ) above may be t e rmed vowe l ' d i s s im i l a t i on ' . S i n c e the vow e l
d i s s imi l at i on o c curs only in c e rt ain w o r d s c on s t r u c t e d by means o f
redup l i c at i on , the ab ove s t ems and the l i ke mu s t b e marke d ' +P-rule 2 5 '
in the le x i c on . It mu s t be n o t e d that vowe l c hanges de s c r i b e d ab ove
may or may not o c c u r on the phone t i c leve l . The re fore , P-ru l e 2 5 s e t
u p b e l ow h a s t o b e regarded a s a n opt ional ru l e .

[
P-rule : Vowe l d i s s im i l ation ( op t i ona l )

� �
J
(25)a. ++ C -conS C (V) ( C ) cons (C) ++ ----+

+Syll +Syll
< +Low> -Low

1
1

2
2

-Cons
3

l
4 5

4 5 6
6

[ � cons
7

8 9 /
8 9

( )
+Sy l l +Syl l

= �� : '
+Low
-

�
h

[ ] [
+Round

J
b. ++ C -con, C (V ) ( C ) con (C) ++ ----+

+Syll +Syl l

[l
< +Round > +Low
1

2 [ �
2

-con'
+Syl l
3

4 5 6
4 5 6

con
+Syll
7

8 9 / 123456789
8 9

< +Low> +High

-Ba c k

where C s t and s for a n y c o n sonant a n d V for a n y vowe l

The f i r s t part o f t he rule ( a ) says t hat :

( a ) a non-low vowe l o c c urring in the final s y l la b l e o f a s t em w i l l b e

phone t i c a l ly r e a l i s e d as [ a ] ( [ -Consonant , + S y l l ab i c , +Low ] ) in
the f i r s t part o f a word c o n struc t e d b y means o f redup l i c at ion
(cf. 5 . 42 ) ;

( b ) and when t he first vow e l i s spe c i fi e d as a low vow e l then i t w i l l

b e phone t i c a l ly r e a l i s e d as [ 0 ] ( [ - C o n s onan t , + S y l l ab i c , -Low ,
-High , +Round ] ) in the f i r s t part o f the deriva t i on ( c f . 5 . 42 : 1- 3 ) .

The s e c ond part o f t he rule ( b ) says t hat

( a ) a low vowe l o c c urri ng in the final s y l lab le of a s t e m w i l l b e

phone t i c a l l y re a l i s e d a s [ i ] ( [ -Con sonant , + S y l l ab i c , +High , -Bac k ] )
126

in the s e c ond part o f a ( t y p i c a l ) word c on s t ru c t e d b y me ans o f re­

dup l i c at i on ( c f . 5 . 43 ) ;

( b ) and when the first vowe l o f the s t em is fur t he r sp e c i f i e d as a round

vowe l , then it wi l l be phone t i c a l l y r e a l i sed as r a J ( [ -Con sonant ,
+Sy l lab i c , +Low J ) in t he s e c ond part o f t he deriva t i on ( c f .
5 . 4 3 : 1- 3 ) .

5 . 9. L OA N W O R D END I NGS

I t was s t at e d i n S e c t ion 1 . 1 . t hat Dut c h and Eng l i s h ( and a l s o

Arab i c ) p lay a s i gn i fi c ant r o l e in t h e deve lopment o f B I , e sp e c i a l ly
in the e n r i c hment o f i t s vocab ulary . In t h i s s e c t i on we are going t o
e xamine a numb er o f doub le t s in words b orrowed from Dut c h and Eng l i sh .
Examp le s ( 5 . 4 4 ) be low show t he a l t e rna t i on be tween t he ( non­
morphemi c ) endings - t a s and - t e t i n a numb e r o f nouns whi c h end in - t y
o r - t e i t i n Eng l i s h and Dut c h respe c t ive ly . H i s t o ri c a l l y speak in g , t he
- t e t ending i s muc h e a r l i e r t han the - t a s . At present however , forms
ending i n - t e t are le s s c ommonly u s e d . Thi s is part ly b e c au s e t he - t e t
ending sounds very Dut c h and part ly b e c au s e B I imp o s e s c o n s t ra in t s o n
the o c c urre n c e o f lei i n word-final s y l l ab l e . ( In nat ive BI i t ems lei
may oc c u r in t he last s y l l ab le b u t t he p r e c e d i n g vow e l mu st b e lei o r
101 . )

( 5 . 4 4 ) I k u a l i t e t l [ k u a l i t a s - k u l i t e t J ' qua H ty '

I k u a n t i t e t l [ k u a n t i a s - k u a n t i t e t J ' qu a n t i t y '
l a k t i f i t e t l [ a k t i f i t a s - a k t i f i t e t J ' a a ti v i ty '
I m a y o r i t e t l [ m a y o r i t a s - m a y o r i t E t J 'majori ty '
I m i n o r i t e t l [ m i n o r i t a s - m i n o r i t E t J 'minori ty '

Examp l e s ( 5 . 4 5 ) b e low show the a l t e rnat ion b e t ween the endings - a l

and - i l in a numb er o f ( c ogna t e ) words whi c h u s ua l l y end in - a l or - e l
in Engl i sh and Dut c h re s p e c t i ve l y . Agai n , the ending - i l is historic­
a l l y muc h e a r l i e r than t h e ending - a l . At pre sent howeve r , t here i s a
s t rong tendency t o sub s t i t u t e - a l for - i l .

( 5 . 4 5 ) I k u l t u r i l l [ k u l t u r a l - k u l t u r L 1 J ' a u l tura l. '

I m a t e r i i l l [ ma t r i a l - m a t e r i : l ] ' ma t e r i a l. '
I s a p i r i t u i l l [ s p i r i t u a l - s p i r i t u L l ] ' sp i ri tua l. '
l i n d i f i d u i l l [ i n d i f i d u a l - i n d i f i d u L l ] ' i n di v i dua l '
I k o n s a t i t u s i o n i l l [ k o n s t i t u s i o n a l - k o n s t i t u s i o n L l ] ' a on s ti tu -
t i o na l. '

The phenomenon i ll u s trated by e xamp l e s ( 5 . 4 4 ) and ( 5 . 4 5 ) above c an

be forma l i s e d as fol lows :
 127

( 26 ) a. X +Cons
-Vo i c e
- C ont
[]
P - ru l e : Loanword endings ( op t iona l )

can
+Sy l l
-High
+Cons
-Vo i c e
- Cont
++ ----+ 1 2 � � con
+Syll
+Low
+Cons
-Vo i c e
+Cont
5

+Ant -Ba c k +Ant +Ant

+Cor +Cor +Cor

[]� 2 3 4 5

� �
J
b. X can c on ++ ----+ 1 con' 3 4
+Sy l l +Son +Sy l l
+High + C ont +Low
-Back

where X s t ands for any s e gment s e quenc e s

Rule ( 2 6 a ) s a y s t hat t he ending I - t e t l i n l oanwords may op t i on a l l y

b e rea l i s e d a s [ - t a s J on the phone t i c leve l ; a n d rule ( 2 6 b ) says t hat
the ending I - i l l in l oanwords may o p t i on a l l y b e r e a l i s e d a s [ - a l J on
the phone t i c leve l .

5 . 10 . STRE S S PLACEMENT

I t has b e e n p o i n t e d out i n s ub s e c t ion 2 . 2 . 5 . that s t re s s i n B I i s

not phonemic b e c au s e i t s p lac ement i n words i s h i ghly pred i c t a b le . As
s t a t e d earl i e r i n the introduc t i on o f t h i s c hapt e r , the mai n c on c e rn
here i s t o a c c ount for s t re s s p la c ement on t he word leve l . Spe c i fi c a l ly ,
t h i s s e c t i on wi l l be c onc erned w i t h t he formu l a t i o n o f rule s t hat w i l l
c o rre c t l y a s s ign s t re s s t o words i n c i t a t i on . At t h i s p o int , i t i s
n e c e s sary t o n o t e t hat s t re s s p lac ement i n a given word may n o t be o n t he
same s y l lab le in a l l c a s e s . There are some i n s t an c e s where syntac t i c a l
fa c t or s , whi c h are n o t our c on c e rn here , may c au s e t he s h i ft o f s t re s s
plac ement i n word s . Pre f i xe s , for examp l e , are b a s i c a l l y un s t r e s s a b l e
on the word leve l . O n the sentence level howeve r , s t r e s s p l a c ement i n
l
word s may b e s h i f t e d t o t he pre f i x e s t o s i gnal gramma t i c a l c ontrast s .
It has b e e n wide ly agre e d by lingu i s t s t hat s t re s s i s a s upra­
s e gmen t a l feature o f u t t e ranc e s w h i c h app l i e s t o whole s y l l ab l e s rather
t han t o individual vowe l s and c on s onant s . Lingui s t i c a l l y s p e ak ing , the
purp o s e o f s t re s s a s s i gnment i s t o mark one s y l lab le and only one

I
For a detailed account of stress placement i n words on the sentence level , and a com­
prehensive review of earlier studies on supra segmental aspects of BI , the reader i s
referred to Halim 197 4 , Chapters 3 and 2 respectively .
128

s y l lab le p e r word ( or per s t re s s uni t ) a s c arrying prominence ( c f .

Hyman 1 9 7 5 : 2 0 4 ; Lade foged 1 9 7 5 : 2 2 2 , e t c . ) .
A s far as prominenc e i s c on c e rne d , t he re i s only one degree o f s t re s s
t o b e re c ogn i s e d in B I words ( in c l uding c omp oun d s and phra s e s ) . All
other un s t r e s s e d s y l lab l e s within a word ( or c ompound or a phra s e ) are
p �onoun c e d w i t h approximate ly the same amount of e n e rgy e a c h .
Let u s o b s e rve some e x amp l e s o f the d i s t r i b u t i o n o f s t re s s in B I
w )rds . Note that i n t he fo l l owing e xamp l e s ( 5 . 4 6 ) e a c h numb e r c o n s i s t s
o f a group o f r e l a t e d i t ems . Note furthermore , that t he a - i t ems b e low
are s tems ( monomorphemic words ) whi l e t he b - and c - i t ems are deriva t i ve s
( oo lymorphemic words ) .

( '5 . 4 6 ) la . / T I K/ [ t fk ] ' to type '

b. / d i + T I K/ [ d i t f k ] ' to be type d '
c. / d i +T I K+ka n + r a / [ d i t i ka n r a ] ' ty p e d b y h i m '
2a . / �AH/ [ sa h , sa h ] ' l e ga l '
b. / t a r+�AH/ [ ta r s a h ] ' t he m o s t l e ga l /au t h e n t i a ( o n e ) '
c. / d i + � A H + k a n +r a / [ d i s a h k a n r a ] , l e ga l i s e d by h i m '
3a . / D U D U K/ [ d ude? ] ' to s i t '
b. / p a+m a n + D U D U K / [ p a n d u d e ? ] ' p opu l a t i o n '
c. / m a n + D U D U K+ i / [ m a n d u d u k i ] ' t o o aaupy '
4a . / T d R I MA/ [ t a dma ] ' to ge t , r e a e i v e (roo t ) '
b. /man+T d R I MA/ [ ma n a r fma ] ' to reae i v e , get '
c. / p a+ m a n +T d R I M A + a n + m u / [ p a n a r i m a ? a n m u ] ' y our aaaep tana e '
5a. / S d l A M A T / [ s a l a ma t ] ' safe '
b. / p a +m a n + S d l A M A T / [ p a r a l a m a t ] ' t he p e rs o n w h o s a v e d '
c. / k a + S d lAMAT+a n / [ ka s a 1 a m a t a n ] ' sa fe ty '

Examp le s ( 5 . 4 6 : 1a , 2a ) are mon o s y l l a b i c and t he re fore t he s t re s s i s

aut oma t i c a l ly a s s i gned t o t hem . The general p a t t e rn of t he p la c ement
o f s t re s s in p o ly s y l l a b i c i t ems shown by t he ab ove e x amp l e s is on the
penult ima t e s y l l a b l e re gard l e s s of whe ther t he penult is part o f a s t e m
or n o t (cf. 5 . 46 : 1c , 2c , 3 , 4 , 5 ) . Examp l e s ( 5 . 4 6 : 1b , 2b ) howeve r , show t hat
the p la c ement o f s t re s s is on t he last s y l lab l e rather t han on t he p e n ­
u l t ima t e s y l l a b l e o f the words . Thi s means that t he p r e f i x e s do not
affe c t the p la c ement o f s t re s s in wo rd s .
Examp l e s ( 5 . 4 7 ) b e low show t ha t t he p l acement o f s t re s s in a word i s
a l s o s h i f t e d from the penult imate s y l lab l e when t he vowe l o f the penult
i E. a s c hwa .

( : . 47 ) 1 . / a m p a t / [ am p a t ] ' four '

/bar i / [ bad ] ' to g i v e '
/ s a r am / [ s a r a m ] ' fri g h t e n i n g '
 129

2 . I s a t a r u l [ s a t a r u ] ' e nemy '

I c a n d a r u l) l [ c a n d a r 6 1) ] ' to b e b ound '
I s a b a r a l) l [ s a b a r a l) ] ' t h e o t h e r s i de , to c r o s s '
3 . I p u t a r i l [ p u t a r i . p u t r i ] ' gi r l , daug h te r '
I ka r a n a l [ ka r ana . karna ] ' b ecaus e '
Icar i taral [ car f tara . car f t ra ] ' s tory '

Not i c e that when a word i s b i s y l la b i c and t he vowe l o f t he f i r s t ( pen­

ult ima t e ) s y l lab le i s . a s c hwa , t he s t re s s will b e a S S i gned t o the final
s y l lab le ( c f . 5 . 4 7 : 1 ) . But when t he word in que s t i on c on s i s t s o f three
or more s y l lable s , and t he vowe l o f t he penult ima t e s y ll ab le is a s c hwa ,
t he s t re s s w i l l be a s s i gn e d t o the antepenult ( c f . 5 . 4 7 : 3 ) provided
that t he vowel o f t he antepenu l t is not a s c hwa in whi c h case the
s t re s s wi l l b e a S S igned t o the final s y l lab le ( c f . 5 . 47 : 2 ) .
Let us now o b s e rve s ome e xamp l e s o f t he p l a c ement o f s t re s s i n
large r un i t s . Spec i fi c a l l y , the f o l lowing e x amp l e s s how t he p la c ement
of s t re s s in redup l i c at ed forms ( c f . 5 . 4 8 : 1 , 2 ) , in c ompound words ( c f .
5 . 4 8 : 3 , 4 ) and in s imp le phra s e s ( c f . 5 . 4 8 : 5 , 6 ) . For c ompar i s on , the
plac ement o f stre s s i n the c omponen t s ( or e lement s ) o f each un i t i s
a l s o given .

( 5 . 4 8 ) 1 . I p a + m a n + m i n t a +m i n t a l [ p a m i n t a m f n t a ] ' b e ggar '

a. Im i n t a l [ m f n t a ] ' to a s k (fo r ) '
b. I p a + m a n +m i n t a l [ p a m f n t a ] ' t h e p e r s o n w h o a s k e d ( fo r ) '
2 . l a n a k + a n a k + l) a l [ a n a ? a n a ? )1 a ] ' hi s / h e r c h i ldren '
a. lanakl [ ana ? ] ' c hi ld '
b. l a n a k +)1 a l [ a n a ? )1 a ] ' h i s/ h e r c h i ld '
3 . I t a n d a ma t a l [ t a n d a m a t a ] ' so u v e n i r , k e e p s ak e '
a. I tandal [ dnda ] ' s i gn , token '
b. Ima t a l [ ma t a ] 'eye '
4 . I r u m a h s a k i t l [ r u m a h s a k L t ] ' ho s p i ta l '
a. I r u m a h l [ r u ma h ] ' ho u s e '
b. Isak i tl [ sakL t ] 'iZZ '
5 . I r u m a h b a r + t i l) k a t l [ r u m a h b a r t f l) k a t ] ' s t or e y e d b u i l d i n g '
a. I ruma h l [ r uma h ] ' ho us e '
b. I b a r + t i l) k a t l [ b a r d l) k a t ] ' ha v i n g s tore y s '
6 . I t u k a l) t i k i [ t u k a l) t f k ] ' ty p i s t '
a . I t u k a n l [ t u k a l) ] ' wo r k man '
b. I t i kl [ t f k ] ' ty p e '

The ab ove e xamp l e s show t hat the p la c e ment o f s t re s s in un i t s t ha t c on­

s i s t o f two or more l e x i c a l forma t i ve s each is on one o f the s y l l ab le s
o f t he l a s t c omponent o f e a c h unit i n a c c ordan c e w i t h the s t re s s p l a c e ­
ment ru l e s i l lu s t ra t e d b y e xamp l e s ( 5 . 4 6 ) and ( 5 . 4 7 ) above .
130

The d i s tribut i on ( p la c ement ) o f s t r e s s o n t he word level i l l u s t r a t e d

by e x amp le s ( 5 . 4 6 ) t hrough ( 5 . 4 8 ) above c a n be summed u p a s f o l l ows :

( i ) The p l a c ement o f s t re s s in words whi c h c on s i s t o f one l e x i c a l

format ive e a c h w i t h o r w i t hout any a f f i x e s added t o t hem w i l l b e
o n t he penu l t imate s y l lab l e e xc ept under t he f o l lowing c i rc um­
s t anc e s :

( a ) the word in que s t i on i s mon o s y l l a b i c whi c h , b y d e f i n i t i on ,

c o n s i s t s o f o n l y one s y l l a b l e and t here fore the s t re s s w i l l b e
aut omat i c a l l y a s s igned t o i t ;

( b ) the vowe l ( nu c l e u s ) o f the penult ima t e s y llab l e o f t he word

in que s t i o n is part of a p r e fi x , in whi c h c a s e t he s t re s s w i l l
b e a s s igned t o t he final s y l lab l e o f t he word ;

( c ) t he vowe l o f the penult imate s y l lab l e o f the word in que s t ion

i s a s c hwa , in whi c h case the s t re s s w i l l b e a s s igned to the
f i n a l s y l lab l e when the word c on s i s t s of two s y l lab l e s only
or the s y l lab l e b e fore t he penult c on t a i n s a s c hwa or be long
t o a pre fi x , and to t he s y l l a b l e be fore t he penult provided
t hat it does not c ontain a s c hwa or b e l on g to a pre fi x .

( i i ) The p l a c ement o f s t re s s in a larger un it ( i . e . a un i t t hat c on­

s i s t s o f two or more l e x i c a l format i v e s w i t h or wit hout any a f f i xe s )
w i l l be in t he l a s t c omp onent o f the unit ( i . e . t he part t hat
c on t a i n s the last l e x i c a l format ive of the unit ) in a c c ordan c e
w i t h r u l e ( i ) above .

Obv i ou s l y , the r e levant c at e gory for t he p la c ement o f s t re s s in B I

i s a phra s e b oundary ( D ) rather t han a fu l l word b o undary . A l t h ough
s t re s s app l i e s to the whol e s y l l ab le , it is a c ommon pra c t i c e in gen­
erative phono logy t o a s s i gn t he s t re s s to t he vow e l of the s t r e s s e d
sy l lab l e . The s t re s s p l a c ement de s c ribed above c an be formul a t e d as
fo l l ows :
 131

�:�:: �
P-rule : Stress placement

(27)a. [ +S t r e s s ] / X ++ C (C) 0
Y (C)

-Round
< +Low>

. . . {C -Cons
o
+Sy l l
+Ba c k
-Round
- Low

b. i- C on sl ----+ [ +S t re s s ] / X ++ ( C
o
-Cons
L:- S Y 1 Jj +Syll
+Ba c k
-Round
-Low

where C s t ands for any c on s onant , V

for any vowe l ,
X and Y c an be e i ther n u l l or non-nu l l , ( ++ ) for a
word b oundary and ( O ) for a phra s e b o undary .

Rule ( 2 7 a ) says t hat s t re s s w i l l b e a s s i gned t o ( a ) t he l a s t vowe l

of a un i t ( phra s e ) i f the l a st c omponent of the unit i s a mon o s y l la b i c
l e x i c a l forma t ive w i t h or w i t hout a n y pre f i x e s added t o i t , a n d ( b ) t o
t h e ( fi r s t ) non - s c hwa vowe l b e fore t h e l a s t vowe l in t he un i t . The
word b oundary ( + + ) prevent s t he ru l e from app lying to p r e f i xe s if t he
l a s t c omponent of t he unit has prefixe s . Rule ( 2 7 b ) says t ha t s t re s s
w i l l be a s s igned t o t he final vowe l o f t he unit i f t he p r e c e d i n g
vowe l { s ) ha s ( have ) the feature spe c i fi c at i on s [ -C on s onant , + S y l l ab i c ,
+Ba c k , -Round , - Low ] . The word b oundary ( ++ ) preve n t s t he rule from
app l y ing t o pre fixe s .

5.11. SUMMARY O F P - RULES

Thi s s e c t i o n pre sent s the summary o f t he phon o l ogi c al ru le s proper

( P-rule s ) s e t up in t he previous s e c t ion s . The ru l e numb e r s given
b e low are the same a s those t hat appear i n the previous s e c t i on s .

( l ) Degemination ( se e sub se c t ion 5 . 1 . 1 . )

+ ----+

whe re C c o n s onant ; Fn a spe c i f i c bun d l e of featur e s

132
(2) � l - - �
Tri l l dele tion ( se e sub se c t ion

con '
o
/ +cv
a
- - - - - ++
5. 1 . 2 .)[ l
con' X
+Son +Son

� r
-Nasal -Na s a l
-Cont -Cont

C -Con s con'
+Syll +Son
+Ba c k -Na s a l
-Round - C ont
-Low

where C any c o n s o nant ; CVa

= / b a - / , / p a - / and / t a - /
=

X a s e quen c e o f s e gmen t s be ginning wi t h a vowe l

=

o -
---
+
f 1
( ] ) Glottal stop insertion ( se e s ub s e c t ion

con,
-Syll
-Vo i c e
/ -Cons
+Syll
aHigh
( + ) -----
5. 2 . 1 .)
-Cons
+Syl l
aHigh
-Cont B Low B Lo w
Round Round

( L, ) Glottal stop rea l i s a tion o f / k / ( se e s ub s e c t i on

+Cons - - -- -.. -Con s / - - - - - ++ ( CVC + )

5. 2 . 2 . )
-Syll -Sy l l
-Vo i c e -Vo i c e
+Bac k +Back
-Low +Low
-Cont - Cont

where C an y c o n s onant ; V any vowe l

 133

[Co-Syllns - - <Va>
( 5 ) Deletion of / h / ( optiona l ) ( s e e s u bsecti o
5.3.1. ) n
+Co-VoicnteJ
.. o / +)
[.-aaLowHigh� ----- [::�:h� ( [::��� ]) ++
+Syll
-Cons +Syll
-Co n s <+Nasal> Verb
Adj
where C any consonant; V
( 6 ) + C [con� [can] V C V ( C( sV)ee ++subsection )
Sy l lable reduc tion ( optiona l ) 5.3.2.

+Syll -Syll
+Low +Low-Cont
1 2 3 4 5 6 7 8 9 10

---- + 1 2 o 0 5 6 7 8 9 10

where C any consonant; V any vowel

( 7 ) Voic ing of / h / ( optiona l ) (
[ -Voice] ----.. [ +Voice] V -Cons ( + ) V
/
s ee s u bsection
5.3.3. )

+Low-Syll
+C o nt
where V any vowel
( 8 ) Place a d j u s tment of / h /

rLo�HiwghJ ----.. j�Hi-LogwhJ /

( s ee s u bsec t i o
5.3.4.)

tLow� n
+Cont aHi+Sygllh
-Syll
-Cons -Cons
1 34

( 9 ) Despiran t i s a tion ( natur a l i sation) of I f I ( optiona l ) ( se e s u b ­

s e c t ion 5 . 4 . 1 . )

[ +Cont inuant ] ----+ [ -Cont inuant ] I

+Cons
+Ant
-Cor
-Vo i c e

( 1 0 ) Natur a l i sation of I z l ( optiona l ) ( se e s ub se c t i on 5 . 4 . 2 . )

a. [ +Vo ic e ] ---- + [ -Vo i c e ] I

+Con s
-Son
+Cont
+Ant
+Cor

[ [
-Sa c k

J J
b. Ant ---- + Ant I
+Cor -Cor
+Cont -Cont
+Cons
+Vo i c e
-Sac k

( 1 1 ) Fronting ( Natura l i sation ) of l s i ( optiona l ) ( s e e sub s e c t i on 5 . 4 . 3 . )

j- Antl ---- +
[+ Antl I
l- CorJ l!- corJ
-Vo i c e
+Cont
-Sa c k
+Cons
 135

f �
( 1 2 ) Natur a l i sa tion of Ixl ( optiona l ) ( s e e s ub s e c t ion 5 . 4 . 4 . )

J J
con, ----+ a. con' I
+Cont -Cont

[
-Vo i c e -Vo i c e
+High +Hi gh
- Sy l l

J
b. con,
-Son
- C ont

[
+Ba c k
+Vo i c e
+High

J
c. con,
+Cont
-Vo i c e
-High

( 1 3 ) Schwa inser tion ( optiona l ) ( s ee sub s e c t i on 5 . 5 . 1 . )

0 - -- - -.. -Cons I [ M E N ] ----- ++CVC ++

+Sy l l
-High
-Low
+Back
-Round

where M E N Im a n l ' ac t ive v o i c e verb pre f ix ' ;

=

++CVC++ mono s y l lab i c s t ems

=
136

[ [
(1 4 ) Nasal dele tion and as s imilation ( s ee sub se c t ion 5 . 5 . 1 . )

J J
con, ----+ a. o / [ M E ] ----- + SY I
+Na s a l +Son
+Ant +Seg
+Cor / [ M E ] ----- + aAnt
b. aAnt

-Syll
a C or a C or
yBack yBack
+Nas a l

[ yll
+Cons -Son
+Seg

J
c. -Ant / [ M E ] ----- + Ant
-Cor -Cor
+Ba c k +S
+Na s a l -Cons
+Con s

where M E the f i r s t two s egment s o f /m a n / j

=

Seg s e gment ( i . e . c on s onant or glide )

=

+ ME [ l
( 1 5 ) Consonant coa le scence ( see sub s e c t i on 5 . 5 . 1 . )

+Cor
Ant

+Na s a l
+ a. +Cons
-Vo i c e
- C ont
x ++ (5 6) (SUFFIX)
+Cons aAnt
aCor
-aBack

b. +Cons
-Vo i c e
+Cont
+Ant
+Cor
1 2 3 4 5 6 7 (5 6) (8)

( cont ' d )
 137

----+ 1 2 c. aAnt 6 7 e. aAnt 6) (8)

BCor BCor
-aBa c k -aBack
+Na s a l +Na s a l
+Cons +Cons

d. -Ant f. -Ant
-Cor -Cor
-Back -Ba c k
+Na s a l +Nas a l
+Cons +Cons
(3 4 5) C3 4 5)

Cond ition : i f a , then c and e ; i f b , then d and f ;

ME the f i r s t two s e gmen t s o f /m a n /
=

X sequence o f s e gmen t s b e ginning w i t h a vowe l

=

[ME] ----+ 0 / + P E + ----- [ l

( 16 ) De letion o f the f i r s t two segments o f /m a n / ( se e sub s e c t ion 5 . 5 . 2 . )

Cons
+Na s a l
+Ant
+ STEM ( SUFF I X )

+Cor

whe re M E the first t wo s e gment s o f /ma n /

PE /pa/

( 1 7 ) Vowe l laxing ( se e s ub s e c t io n 5 . 6 . 1 . )

r- C onsl ----+ [ -Ten s e ] / [ - -- - - ] C ++

I S Y l lJ
�

where C any c on s onant

( 1 8 ) Vowe l re trac tion ( optiona l ) ( se e sub s e c t ion 5 . 6 . 2 . )

[ -Bac k ] ----+ [ +Bac k ] / X �

C VC VC ( C VC ) ++
o o o o
-High
-Low
-Round
+Syll
-Cons
- S t re s s
where X any s e gment se quenc e s o r nul l
138

( 1 9 ) Schwa d e l e ti on ( optiona l ) ( s ee sub s e c t i on 5 . 6 . 3 . )

-Cons
+Syll
-Low
+Ba c k
-Round

where C any c on s onant ; V any vowe l ;

=

X any se quenc e o f se gment s or n u l l

( 2 0 ) Vowe l na s a l i sation ( optiona l ) ( se e s ub s e c t ion 5 . 6 . 4 . )

r+ C on s
J ��� � � �Jl
[ -Nasa l ] ---- + [ +Na s a l ] / X y
li Na s a l
con

r�r �
+ ���� ��� +
-Cons -Cons
I n Arab i c
+Low +Low -
words
where X , Y any seque n c e o f s e gmen t s

( 2 1 ) Vowe l lowering ( optional ) ( s ee sub se c t i on 5 . 6 . 5 . )

a. [ +H i gh ] - - - -+ [ -H i gh ] /
-Low
aRound
aBa c k
+Sy l l
-Cons I n a number o f it ems

b. [ -Low ] ----+ [ +Low ] /

-High
+Ba c k
-Round
+Syll
-Cons In a number o f i t em s
 139

( 2 2 ) Vowe l coale scence ( optiona l )

++ X [LOW� [Hlg� C V ( C ) (
++ s ee
----+
s u bsecti o n
5.6.6. )

+Back
+Syll +Bac+Syllk
1 2
-Cons -Cons
3 4

1 2 -Hi-Lowgh 5 6 7 8

+Round
+Syll
-Cons (3 4)

( 2 3 ) G l ide inser tion ( optiona l ) ( s ee s u bsection 5.7.1. )

+High J <-aRound>, J
[Cons
-Syll l X �cons /
+Syll ----- [ <+Cons ::��� y

+Voice aRound
aRound
where X, any segment sequence or null
Y

[+SyllJ [ -SyllJ {���a} [-�:�:] [ <:���:

( 2 4 ) Desyllabica tion ( optional )

/
( s e e s u bsecti o n
5.7.2. )

+High>cJ
----+

++ aRound +High <-aRound>

Condi tion :
CXif a,anyanythencsegonsmb;oentnant;ifsec,quencVtheneansydorvowel;null
=
=
 ++ C [ �
( �� 5 ) Vowe l d i s s im i l a tion ( option a l )

a. -con
+Syl l
C (V ) ( C ) [ �
con
+Syll
( s e e s e c t ion 5 . 8 . )

( C ) ++ - - - -..

< +Lo w > -Low

1 2 3 4 5 6

[ �
7 8 9

CO W )
1 2 -Cons 4 5 6 con 8 9 / ----- 2 34 5 6 7 8 9
+Syl l +Sy l l
+Low
- H i gh

[ ] [ �
+Round

b. ++ C -cons C (V) ( C ) con ( C ) ++ ----+

+Syll +Syll

[]
< +Round> +Low
1

2 [ �
2

-con
+Syll
3

4 5 6
4 5 6

con
+Syll
7 8

8 9 / 123456789
9

< +Lo w> +Hi gh

-Back

whe re C any c on s onant ; V any vow e l

+Cons
-Vo i c e
- C ont
[]
( 2 6 ) Loanword endings ( optiona l )

a. X con
+Syll
-H i gh
+Cons
-Vo i c e
-Cont
++ ----+ 1 2 [ �
( see s e c t i on 5 . 9 . )

con
+Syl l
+Low
+Cons
-Vo i c e
+Cont
5

+Ant -Back +An t +An t

+Cor +Cor +Cor

� [ �
1 2 3 4

b. X

r] +Sy l l
+High
-Back
n con
+Son
+Cont J
++ - - - - ... 1 con
+Sy l l
+Low
3 4

1 2 3 4

whe re X s t an d s for a n y s e gment s equenc e s

 141

( 2 7 ) Stress placement ( s ee s e c t i on 5 . 1 0 . )

a. i- C onsl ----+ [ +S t re s s ] / X ++ C (C) 0

I : ::::
li S Y l lJ

J
Y (C)

� +LOW>
d

. . . (C -Cons
o
+Syll
+Ba c k
-Round
-Low

b. i- Consl ---- + [ +S t re s s ] / X ++ ( C
o
-Cons o
li Sy l lJ +Syll
+Ba c k
-Round
-Low

where C any c on s onant ; V

= any vowe l ; X , Y
= any =

seque n c e of s e gme n t s o r nul l ; 0 phrase b oundary

=
 CHAPTER 6

SUMMARY AND CON CLUSIONS

It was s t at e d in Chap t e r 1 t hat t he primary goal of t h i s s t udy i s

t o pre sent a phono l o g i c a l de s c ript ion o f B I wi thin a t rans format ional
generative framework . The de s c ri p t ion p r e s e n t ed i n t he s t udy c entres
around ( 1 ) t he pro b l em o f the repre sent at i on of BI morpheme s both on
the phonologi c a l and phone t i c leve l s ( c f . Chap t e r 2 ) , ( 2 ) the prob lem
of the range of p o s s i b l e ( or we l l - formed ) morpheme s and s e gment s i n BI
( c f . Chap t e r s 3 and 4 ) , and ( 3 ) t he problem of phone t i c forms of B I
morpheme s whi c h , under c e r t a i n c ir c umstanc e s , b e c ome d i ffe rent from
the i r r e sp e c t ive phon o l o g i c a l ( unde r l y i n g ) forms ( c f . C hap t e r 5 ) .
We saw in Chapter 2 that B I require s 2 2 c on sonan t a l ( non-s y l lab i c )
s e gment s ( i . e . 1 8 nat ive and 4 l oan s e gment s ) and s i x vowe l s for the
repre s e n t at i on o f morpheme s on the phono l o g i c a l leve l . I t was argued
that the glot t a l s t op ( [ 1 J ) is not re levant on the phono l o g i c a l l e v e l
o f repre sent a t i on , d e s p i t e the fac t t hat the s e gment i s phonemi c i n
trad i t i onal ( o r aut onomou s ) phonemic anal y s i s . I t was demon s t rated
t hat [ 1 J i s b a s i c a l l y a phone t ic real i sa t i on of t he unde r l y i ng / k / in
morpheme final p o s i t i on . ( When in morpheme m i d p o s i t ion , [ 1 J fun c t i on s
t o b reak u p s e quenc e s o f i dent i c a l vowe l s on t he phone t i c leve l . In
other words , [ 1 J in morpheme m i d p o s i t ion i s phono l o gi c a ll y repre s e n t e d
b y zero . ) I t was p o i n t e d o u t t hat t he velar s t op ( [ k J ) real i sa t i o n o f
final / k / i n words l ike / p a k / [ p a k J ' p a c kage ' , / p o l i t i k/ [ p o l i t l k J
'po li tics ' , / f a n a t i k/ [ f a n a t l k J ' fan a t i c ' , e t c . , whi c h is n o t o b s e rved
i n native BI i t ems can b e a c c ounted for by mak i n g use of the hi s t o r i c a l
i nforma t i on about the i t ems . The h i s t o ri c a l ( e t ymo logi c a l ) i n format ion
also explains the free var iab i l i t y be tween c e rt a i n s o und s , s u c h a s [ P J
and [ f J i n words l ike p i k i r ' to t h i n k ' , p a h am ' t o comp r e h e n d , i d e o l o gy ' ,
p i ha k ' s i de ' , e t c . , [ z J and [ J J �r [ s J i n word s l ike z a m a n ' e ra ' ,
i j azah ' di p l o ma ' , a z a s 'princip le ' , etc . , [ s J and [ s J i n words l i k e

143
syah ' le ga l ' , sya r a t ' re quiremen t s ' , ma s y a r a k a t ' s oc i e ty ' , e t c . , [x]
and [ k ] i n words l ike k h a b a r 'news ' , khas ' ty p i ca l ' , a k h i r ' end ' , etc .
It was c onc luded t hat such ( l oan ) i t ems have t o be phonologi c a l ly rep­
re s e n t e d w i t h the l oan s e gme n t s irre spe c t ive o f t he i r h i s t ory i n t he
1 3.nguage . Thus the i t ems p i k i r [ p i k L r , fikLr] ' to think ' , j ama n / zam a n
[ l a ma n , z a m a n , s a m a n ] ' e ra ' , m a s a r a k a t / m a s y a r a k a t [ m a s a r a k a t , m a s a r a k a t ]
' s o c i e ty ' , k h a b a r/ka r a b [ ka ba r , xa ba r , h a b a r ] ' n e w s ' , e t c . are re spe c ­
t ive l y rep r e s e nt e d a s I f i k i r / , I z a m a n / , Ima s a r a k a t l and Ix a b a r l o n t he
p hono l og i c a l leve l d e s p i t e t he fac t that t he s e i tems do not s ound
fore i gn to many spe ake r s of BI in t he s e n s e t hat t hey are a lways p r o ­
noun c e d w i t h t he nat ive s ound sub st it ut e s . And probab l y many speakers
do not know t hat they are l oanword s . The mot ivat i on behind t h i s ap­
p r o a c h is to make the grammar simp l e .
We saw , furthermore , t ha t t he 22 c on sonant a l s e gment s and the s i x
v o we l s c a n be ke pt d i s t i n c t from one anothe r by making u s e o f 1 2
s i f i c a t ory ) feature s . I t w a s s t a t e d t hat s t op s and a f fr i c a t e s a r e n o t
( c la s ­

d i f fe r en t i a t e d in t e rms o f r e l e a s e feature s b e c au s e i n B I s t op s ( / p ,
b, t , d , k , g / ) are a l l non-p a l a t a l c on sonant s , whi le a f fr i c at e s ( / c ,
l / ) are b o t h p a l a t a l s ounds . In o t h e r words , the p l a c e features
c.nter ior , corona l and back are s u f f i c ient to keep stops and a ffri c a t e s
apar t . A s f a r a s t he nat ive speake r s o f B I a r e c oncerne d , t he feat ure
spe c i f i c a t i on s [
for s t o p s are redundant .
+ Delayed re lease ] for a f fr i c a te s and
From the point o f view o f t he l ingui s t , who
[ - Delayed r e l e a s e ]
has a far more c omp l i c a t e d sound sy s t e m , the r e l e a s e feature s are
c e r t a i n l y c ru c i a l in order to d i f fe rent i a t e , for e xamp l e , the v o i c e l e s s
:_ab i a l s t op [ p ] from t he c orresponding a f f r i c a t e [ p ] or t he v o i c e l e s s
}
p a l a t a l a f fr i c ate [ c ] ( = I PA [ t Q ] ) from t h e c orre sponding s t op [ c ] , e t c .
�Phe grammar howeve r , i s intended as a d e s c r i p t i on o f the c omp e t e n c e o f
131 nat ive speake r s . There fore , one find s no need t o d i f feren t i at e
s t o p s from a f fr i c at e s b y re l e a se feat ure s .
Chap t e r 3 p r e s e n t s the spe c i fi c a t i o n s o f we l l - formed morphemes i n
131 in t erms o f ' po s i t i ve c on d i t i on s ' . It w a s s t a t e d , among o t he r
': hings , t hat a c on s onant c lu s t e r in B I m a y c on s i s t , on t he b a s i s o f
the ob s e rvat ion o f nat ive B I i t ems , o f t w o c on s onants only a n d i t may
o c c ur i n morpheme -me d i a l p o s i t i on only . It was a l s o s t at e d that t he
po s i t ive c ondi t i on s hold t rue mainly w i t h nat ive B I morpheme s . It was
n o t e d t ha t l oan i t ems whi c h d o not mee t t h o s e p o s i t ive c on d i t i on s must
b e repre s e n t e d i n such a way on t he phono l o g i c a l l e ve l , i . e . b y in­
sert ing s c hwa , s o that t h e i r underly ing forms sat i s fy t ho s e c ondi t i on s .
rhu s ( )
[struktor]
l oan i t ems l ike k l i n i k [ k l i n L k ,
' s truc ture ' , i ns truks i
kel i n L k ]
[ i n s t ruks i ]
' c linic ' , s t ruktur
' i n s tru c t i o n ' , e t c . mu s t
 145

re s p e c t ively b e repre s e n t e d o n the phono l o g i c a l l e v e l a s / k a l i n i k / ,

/ s a t r u k t u r / and / i n s a t r u k s i / . The i n s e r t e d s c hwa i s l a t e r d e l e t e d by
a phone t i c rea l i sa t i on rule ( c f . P-rule 1 9 ) . The p o s tulat i on of s c hwa
( /a / ) in the unde rlying repre s en t a t i o n of the above l oanwords is art i ­
f i c i a l in t he s e n se that educated speak er s never real i se t h e s c hwa on
the phon e t i c l e ve l . Obviou s l y , t h i s phenomenon n e e d s a further inve s ­
t igat ion .
In addi t i on t o the p o s i t ive c on di t i on s , Chap t e r 3 a l s o pre sent s t he
d i s c u s s i on o f a c ertain type o f words ( derivat ive s ) wh i c h are u s u a l l y
ac c ounted f o r in an ad hoc w a y in earl i e r s t ud ie s . It was demon s t r a t e d
t ha t the re l a t e d p a i r s ( i . e . forms d e r i v e d from the same s t e m ) s u c h a s
p e k e r j a [ p a k a r J a ] ' wo rkman ' vs . p e n g e r j a [ p a l) a r J a ] ' t h e e x e c u t o r of a
job ' , pese r t a [ pas a r ta ] 'par t i c ipan t ' vs . penye r t a [ papa r t a ] 'a c ompany
( fr i e n d ) ' , p e s u r u h [ p a s u r u h ] ' a j a n i tor ' vs . p e n y u r u h [ p a p u r u h ] ' t h e
p e r s o n who gave order ' , e t c . c an be e x p lained in a phono l o gi c a l ly
p la u s i b l e way by e s t a b l i shing t wo d i f ferent t y p e s o f underlying s t ruc ­
t ure , name l y ( 1 ) P E + STEM ( for the f i r s t s e t of the i t ems ) , and
(2) PE + M E N + STEM ( for the s e c ond set of the i tems ) . The anal y s i s
exp l a i n s why s t em- i n i t i a l c on s onants i n t h e f i r s t s e t o f t he i t ems d o
n o t undergo any c hange when pre ceded by t he p r e f i x P E ( /pa/) ' agent
noun forming pre f i x ' as t he y do in t he se c on d s e t . In t h i s r e s p e c t , a
t ra n s f ormat ional generat ive analy s i s i s c learly mo re powerful t han a
s truc tura li sm-based anal y s i s .
In Chap t e r 4 t he que s t i on o f the morpheme we l l - formedn e s s i s furt her
d i s c u s se d . It was pointed out , among other things , that B I imp o s e s
c on s traint s o n t h e t ype o f c on sonan t s t hat may o c c ur i n a s e quen c e .
Voi c e d c on sonan t s t hat may oc cur as t he first e lement o f a c on s onant
c lu s t er , for e xamp le , are very limited . We saw that only s onorant
c on sonant s ( n a s a l s and l i qui d s ) may prec ede v o i c e d a s we l l a s vo i c e le s s
c o n s onant s . We a l s o saw t ha t v o i c e d c on s onan t s that may o c c ur i n
morpheme - final p o s i t i o n are limited t o s onorant c o n s onant s . The l a s t ­
men t i oned cond i t i o n i s s t rong in t he s e n s e t h a t a loan i t em ending i n
a non-s onorant v o i c e d c on s onant in t h e source language a lways e n d s i n
t h e c orre sponding homorgan i c vo i c e l e s s c on sonant i n B I s u c h a s s e b a b
[ s a b a p ] ' b e ca u s e , cau s e ' ( Arab i c ) , A h a d [ a h a t ] ' Sunda y ' ( Arab i c ) , k l u b
[ k l o p , k a l o p ] ' c l u b ' ( Dut c h / Eng l i s h ) , e t c . Not i c e that the ini t i a l
c on s onant c l u s t e r i n k l u b may b e pre serve d on t he phon e t i c leve l . In
o t her words , t he c on s t raint on t he init i a l c on s onant c lu s t e r i s l e s s
s t rong c ompared w i t h t h e c on s t raint o n c onsonan t s i n morpheme - f i n a l
p o s i t ion .
146

I t was a l s o shown that b y making u s e o f the hi s t or i c a l informa t i on

ab out word s , t rue lingui s t i c genera l i s a t i o n s c an be made to the e xt e n t
t h a t c ommon morp heme s t ruc ture s ( i . e . s t r u c t u r e s whi c h o c c ur in nat ive
B1 i t e ms and to s ome degree in l oan it ems ) c an b e di st ingu i s he d from
unc ommon one s ( i . e . structur e s whi c h o c c ur in loanword s only ) .
In Chapter 5 t he d i s c u s s ion was d i re c t e d t o t he formulat ion o f
phono logi c a l rule s proper ( P-rule s ) or phon e t i c rea l i s at i on ru l e s .
We saw t hat B I ' St andard Pronun c iat ion ' require s about 3 0 P-ru l e s t o
a e c ount f o r variat ions in shape t ha t morpheme s unde rgo in various en­
vi ronment s .
As stated in S e c t i on 1 . 2 . , t h i s study i s an int rodu c t i o n t o t he
phono l o gy o f B I . As an i n t roduc t i on , it invi t e s furt her r e s e ar c h t o
c on firm , sup p le ment and e ven t o re fute t he findings reported in t h i s
s t udy .
 B I B L I O G R A PH Y

ABERC ROMBIE , David

1 9 56 P� o b l e m6 a n d P�i n ciple6 : S tudi e6 i n t h e T ea c hi n g 06

E n gli6 h a6 a S e c o n d L a ng uag e . London : Longman s , Green .

1 967 E l e m e n t6 0 6 G e n e�al P h o n etic6 . Ed inburgh : Edinburgh

Unive r s i t y P re s s .

AINSWORTH , W . A .
1 9 76 M e c hani6 m6 0 6 S p e e c h R e c o g ni ti o n . Ox ford : Pergamon Pre s s .

ALI SJAHBANA , S . Takd i r

1948 Tata b a h a6 a Ba�u B a h a6 a I n d o n e6 ia ( A New Grammar o f Bahasa
Indone s i a ) 2 vol s . Jakart a : P u s t aka Rakyat . 4th p r i n t i n g ,
195 1 .

1962 I n d o n e6 � a n L a n g uag e a n d L�te�atu� e : T w o E 6 6 a y6 . C u l tural

Report No . l l . New Haven : Yale Unive r s i t y S outhe ast A s i a
Studi e s .

1971 ' Some P lanning Proc e s se s i n t he Deve lopment o f the

Indon e s i an-Ma lay Language ' . In J oan Rub in and Bj orn H .
Je rnudd , eds Can L a n g u a g e b e Pla n n e d ? , 179-187 . Hono l u l u :
The Unive r s i t y Pre s s o f Hawa i i .

ANDERSON , S tephen R .
1972 ' The Faro e s e Vowe l S y s t em ' . I n Brame , ed . 1 9 7 2 : 1- 2 1 .

1974 T h e O�g ani z ati o n 06 Phonolog y . New York : A c ademi c Pre s s .

14 7
148

BOTHA , Rudol f P .
1968 T h e F u n Qt� o n 0 6 t h e L e x�Q o n i n T�anh 6 0 �mati o nal G e n e�at� v e

G�a mma�. The Hague : Mouton .

BHAME , Michael K .
1972a ' The S e gme n t a l C y c l e ' . In Brame , e d . 1972 : 62-72 .

1 9 72b ' On the Ab s t rac t n e s s o f Phono l o gy : Ma l t e s e " . In Brame ,

ed . 1972 : 2 2-61 .

B RAME , MichaeI K . , ed .
1972 C o nt�� b uti o nh t o G e n e�ati v e P h o n o l o g y . Au s t i n : Unive r s i t y
o f Texas Pre s s .

BHANDSTETTE R , Renward
1916 A n I n t�o duQti o n t o I n d o n eh i a n L � n g uih t� Q h . Tran s la t e d i n t o
Eng l i s h b y C . O . B lagden . Monograph 1 5 . London : Royal
A s i at i c S o c i et y .

B ROWN , G i l l ian
1972 P h o n o l o g � Qal R u l e h a n d Vial e Q t V a�ia� o n : A Study 0 6 the

Ph o n o l o g y 0 6 L um a h aa b a . Camb ri dge : Cambri dge Unive r s i t y

Pre s s .

CATFORD , J . C .
1968 ' The Art i c ulat ory Po s s i b i l i t i e s o f Man ' . I n Bert i l
Malmb e r g , ed . M a n ual 0 6 P h o n etiQh , 309-333 . Ams t e rdam :
North-Ho l land .

1977 F u n d a m e ntal P� o b l emh i n P lt o n et� Q h . Edinb urgh : Edinburgh

Unive r s i t y Pre s s .

CHEN , Ma tthew
1973 ' On t he Formal Expre s s ion o f Natural Rul e s i n Phono logy ' .
J o u�nal 0 6 L � n g uih tiQh 9 : 2 2 3 -2 4 9 .

CHOMSKY , Noam
1 9 57 S y nta Q ti Q St�UQtu�eh . The Hague : Mout on .
 149

1962 ' A Tran s fo rma t i o n a l Approac h t o Syntax ' . I n A . A . Hi l l ,

ed . P�o c e e d i n g � 0 6 t h e T hi�d T e xa� C o n 6 e� e n c e o n P�o b l em�

06 L i n g ui � ti c A naly�i� i n E n g li� h , 1 95 8 , 124-158 . Re­

printed in Fodor & Kat z , e d s , 1 9 6 4 : 2 1 1- 2 4 5 .

1964 C u�� ent I � � u e� i n L i n g ui� ti c T h e o � y. The Hague : Mouton .

1965 A� p e c t� 0 6 t h e T h e o � y 0 6 S y ntax. Camb r i d ge , Ma s s . : M . I . T .

Pre s s .

1966 T o pi c � in t h e T h e o � y 0 6 G e n e�ati v e G�amma� . The Hague :

Mout on .

CHOMSKY , Noam and Mor r i s HALLE

1968 T h e S o u nd Patte�n 0 6 E n g li� h . New York : Harp e r & Row .

CHOMSKY , Noam and George A . MULLER

1963 ' Introduc t i on t o t he Formal Ana l y s i s o f Natural Language s ' .
In R . D . Luc e , R . R . Bush and E . Galant e r , e d s Handb o o k 0 6

Math emati c al P� y c h o lo g y , 269-32 1 . New York : Wi l e y .

CRYSTAL , David
1969 P � o � o dic S y� t em� a n d I n t o n a ti o n i n E n g li� h . Camb r i dge :
Camb ridge Unive r s i t y Pre s s .

DARDJOWIDJOJO , Soen j ono

1966 Indone s ian Syntax . Ph . D . d i s se rt a t i on , Ge orget own
Univer s i t y , Wa shingt o n .

DERWING , Bruce L .
1973 T�an� 6 0 �mati o nal G�amma� a � a T h e o � y 0 6 L a n g uag e A c q ui� i ­
ti o n . Cambridge : Camb ri dge Univers i t y Pre s s .

DEW , Dona l d and Paul J . JENSEN

1977 P h o n eti c P�o c e� � i n g . Co lumb u s , Ohio : Be l l & Howe l l .

FANT , Gunnar
1973 S p e e c h S o u n d� a n d F eatu�e�. Camb ri dge , Ma s s . : M . I . T .
Pre s s .
150

F I SCHE R-J�RGENSEN , E l i
1975 T�e nd� i n P h o n o lo g i cal T h e o� y . Copenhagen : Akademi sk
Forlag .

F ODOR , Jerry A . and Jerrold J . KATZ , eds

1964 T h e S��uc�u�e 06 L a n g uag e : R e a d i n g � i n � h e Philo � o p h y 0 6

L ang uag e . Englewood C l i f f s , N . J . : Prent i c e - Ha l l .

F RASE R , John F .
1892 ' On t he Language s of Oceania ' . J o u�nal 0 6 � h e R o yal

S o ci e� y 0 6 N ew S o u� h Wa!e� 26 : 342-362 .

f!ALIM , Amran
1972 ' Edj aan Jang D i s empurnakan dan Perkembangan I lmu Bahas a ' .
( Re forme d Spe l li n g and Lingui s t i c Deve l opmen t ) . In
Dj ayanto Supra a n d Ant on Lake , e d s S e mina� B a ha� a

I n d o n e� ia 1972, 1 4 2 - 15 7 . ( Seminar on Baha sa Indone s i a ,

1972 ( report ) . ) Ende h , F l o re s : Nu sa Indah .

1974 I n � o n a �i o n i n R ela�i o n � o S yn�a x i n B a h a� a I n d o n e� i a .

Jakart a : Lembaga Bahasa Nas i onal & Pene r b i t Dj ambat an .

1976 ' Fungsi dan Kedudukan Baha s a Indone s i a ' . ( The Fun c t i on
and S t a t u s o f Baha sa Indone s ia ) . In Ha l i m , ed . 1 9 7 6 , vo l . 2 :
17-26 .

f�LIM , Amran , ed .
1976 P o li�i k B a ha� a N a� i o n a! . ( Na t i onal Language P o l i c y ; Report
on Seminar 2 5 - 2 8 Fe bruary , 1 9 7 5 ) . 2 vols . Jakart a :
Lembaga Bahasa N a s i ona l .

HALLE , Mor r i s
1959 The Sound Pa��e�n 0 6 R u � � ian. The Hague : Mou t on

1962 ' Phono logy i n Generat i ve Grammar ' . Wo�d 1 8 : 5 4 -7 2 . Re ­

p r i n t ed in Fodor and Kat z , eds , 1 9 6 4 : 3 3 4 - 3 5 2 .

1964 ' On t he Ba s e s o f Phono logy ' . I n Fodor and Kat z , e d s 1 9 6 4 :

324-333 .
 151

1973 ' Pr o l e gomena t o a Theory o f Word Format i on ' . L i n g uihzic

I n q ui�y 4 / 1 : 3 - 1 6 .

HARMS , Robert T .
1968 I nz� o d u c zi o n z o P h o n o l o g ical T h e o � y . Englewood C l i f f s ,
N . J . : Pre n t i c e -Hal l .

HARTMANN , R . R . K . and F . C . STORK

1972 Viczi o n a�y o n L a n g uag e a n d L i n g uih zi c h . London : App l i e d
S c i en c e Pub l i s her s .

HENDERSON , Eug e n i e J . A .
1971 ' Struc tura l Organ i z a t i o n o f Language I - Phono l o gy ' . In
Noe l Minni s , e d . L i n g uih zich a z L a �g e , 35-5 3 . London :
G o l l anc z . Reprinted 1 9 7 3 , S t . Albans , Hert ford s h i re :
Paladin .

HENDERSON , Eug e nie J . A . , e d .

1971 T h e I n dih p e n h a b l e F o undazi o n : A S el e czi o n n � o m z h e W�izi n g h
on H e n � y Swe ez . London : Ox ford Unive r s i t y Pre s s .

HOARD , Jame s E .
1972 ' Na t uralne s s Cond i t ion in Phono logy , w i t h Part i c ular
Re feren c e t o Engl i s h Vowe l s ' . In Brame , e d . 1972 : 12 3 - 1 5 4 .

HOCKETT , Cha r l e s F .
1955 A M a n ual o n P h o n o lo g y . Balt imore : Waverly Pre s s .

HOOPER , Joan B .
1976 An I n z� o duczi o n Z O Nazu�al G e n e�azi v e P h o n o l o g y . New
York : A c ademic Pre s s .

HORNE , El inor C .
1961 B e g i n n i n g J a v a n eh e . New Haven : Y a l e Univers i t y Pre s s .

HYMAN , Larry M .
1975 Phonolog y : T h e o � y a n d A na l Yh ih . New York : H o l t , Rine hart
and Win s t on .

INDONE S IA : Panitia Edj aan Lembaga Bahasa dan Kesus a s teraan

1966 E d j a a n Ba�u B a h a h a I n do n e h ia . ( The New Spe l l ing o f Baha s a
Indone s i a ) . Jakart a : Dian Rakya t .
152

I SKANDAR , N .
1975 ' When Z . P . G . i n Indone s ia ? ' . T h e I � d o � e� i a � Qua�t e�t y

3/3 : 3- 3 0 .

JAKOB SON , Roman , C . Gunnar M . FANT and Mor r i s HALLE

1952 P�etimi�a�ie� t o S p e e e h A � a t y � i � . 1 0 t h print ing , 1 9 7 2 .
Cambridge , Ma s s . : M . I . T . Pre s s .

Jl\KOBSON , Roman and Mor r i s HALLE

1956 F u �dam e�tat� 06 L a � 9 uag e. The Hague : Mouton . Rev i se d
e d i t i on , 1 9 7 5 .

JOHNS , Yohanni
1975 B a ha � a I �d o � e� ia : L a �9 k a h Ba�u : A N ew A p p�oa e h. Canberra :
Department o f Indone s i an Language s and L i t e rature s ,
A u s t r a l i an Nat i on a l Unive r s i t y .

JONES , Dan i e l
1950 T h e P h o � em e : I t� Natu�e a � d U� e . Camb r i dge : W . Heffer .
Third e d i t ion 1 9 7 6 : Camb ridge Unive r s i t y Pre s s .

Ki:�HLER , Hans
1956 G�ammatik de� Baha� a I � d o � e� ia. Wi e s baden : Ot t o
Harra s s owi t z .

KE:NSTOWIC Z , Micha e l J . and Charles W . K I SSEBERTH

1977 T o pie� i � P h o � o t 0 9 i eat T h e o �y. New York : Ac ademic Pre s s .

K I M , Chin-Wu
1972 ' Two Phono logi c a l Not e s : A- sharp and B - flat ' . In Brame ,
ed . 1972 : 155-170 .

K I NG , Robert D .
196 9 Hi� t o�ieat L i � 9 ui� tie� a � d G e � e4ati v e G�amma�. Engl ewood
C l i ffs , N . J . : Pren t i c e - Ha l l .

K I PARSKY , Paul
1972 ' Exp lanat i on i n Phono l ogy ' . In S t a n l e y Pete rs , ed . G o at�

06 L i � 9 ui � ti e T h e o � y , 189-2 2 7 . Englewood C l i f f s , N . J . :
Prent i c e -Hal l .
 153

1973 ' '' E l s ewhe re '' i n Phonol ogy ' . I n Ander s on and Kiparsky ,
eds , 197 3 : 9 3-106 .

K I SSEBERTH , Char les W .

1972 ' On Deriva t i ve Propert i e s o f Phon o l o gi c a l Rul e s ' . In
Brame , e d . 1972 : 201-228 .

LADEFOGED , P eter
1971 P � e timina�ie¢ �o L i n g ui¢ �i c P h o n e�i c¢ . C h i c ago : The
Univer s i t y of C h i c ago Pre s s .

1975 A C o u�¢ e i n P h o n e�ic¢ . New York : Harc ourt , Bra c e ,

Jovanovi c h .

LAS S , Roger
1976 E n g ti¢ h P h o n o to g y a n d P h o n o t o g i ca t T h e o � y : S y n c h� o ni c a n d

Viac h�o nic S �u d i e¢ . Cambridge : Cambr i dge Unive r s i t y Pre s s .

LEH I STE , I l se
1970 S u p�a¢ eg m en�at¢ . Camb ridge , Ma s s . : M . I . T . Pre s s .

LYONS , John
1962 ' Phonemi c and Non-phonem i c Phonol ogy : S o me Typ o l o gi c a l
Reflections ' . I n �e�na�i o na t J o u�nat 0 6 Am e�i c a n L i n g ui¢ �i c¢
28/2 : 127-13 3 . Repr i n t e d i n W . E . Jones and J . Lave r , e d s ,
197 3 , P h o n e�ic¢ in L i n g ui¢ �i c¢ : A Boo� 0 6 R e adin g ¢ , 229-239 .
London : Longman .

1968 I n �� o d u c �i o n � o T h e o � e�i cat L i n g ui ¢ �i c ¢ . Camb ridge :

Cambri dge Uni ve r s i t y Pre s s .

MACDONALD , R . Ro ss and Soe n j ono DARDJOWIDJOJO

1967 A S � u d e n� ' ¢ R e 6 e� e n c e G�amma� 0 6 M o d e � n F o �mat I n d o n e¢ ia n .

Wa s h ington , D . C . : G eo rget own Unive r s i t y Pre s s .

MATTHEWS , P . H .
1972 I n 6 t e c �i o nat M o � p h o t o g y : A T h e o � e�icat S � u d y B a¢ e d o n
I
A ¢ p e c �¢ 06 L a�in V e� b C o n j ug a�i o n. Cambr idge : C amb r i d ge
Univer s i t y Pre s s .
l �) 4

1974 M o �pholo g q : A n I n�� o d uc�� o n � o � h e T h e o � q 0 6 W o � d - S �� u c�u� e .

Lond on : Cambridge Univer s i t y Pre s s .

MULDER , J . W . F .
1968 S e�� a n d R ela�� o n� � n P h o n o lo g q : A n A x� o m a��c A p p � o a c h �o

� h e V e� c��p�� o n 0 6 S p e e c h. London : C larendon Pre s s .

P IKE , K . L .
1943 P h o n e��c� . Ann Arbor : The Unive r s i t y o f M i c h i gan Pre s s .
1 0 t h printing , 1 9 7 2 .

1947 ' Grammat i c a l Prerequ i s i t e s t o Phonemic Ana l y s i s ' . Wo�d 3:

155-172 . Reprin t e d i n Erik C . Fudge , ed . 197 3 , Phonolog q :
S e l e c � e d R e ad� n g � . Harmond swort h , Middl e s e x : Pengu i n
Books .

ROS I D I , Aj ip
1976 ' Pengembangan Baha sa Daerah ' . ( Re g i ona l Language Deve l op ­
ment ) . In Hal i m , e d . 1 9 7 6 , vo l . 2 : 1 0 1 - 1 1 5 .

RUWET , Nico l a s
1973 An I n �� o du c �� o n � o G e n e�a�� v e G�amma� . Tran s la t e d i n t o
Engl i s h by Norval S . H . Smi t h . Ams t e rdam : North-Ho l land .

SAMSURI
1960 ' Si s t im Fonem Indone s i a dan s ua t u Penj u s unan Edj aan Baru ' .
( Sy s t em of Indone s i an Phoneme s and the Formu l a t i on of a
N e w S p e l ling . ) M e dan I lm u P e n g e�ahuan 1 / 4 : 3 2 3 - 3 4 1 .

SCHACHTER , Paul and Vic toria FROMK IN

1968 A P h o n o lo g q 0 6 A �a n : A � ua p e m , A� a n � e , and Fan�e . Working
Pape r s in Phon e t i c s 9 . Lo s Ange le s : Unive r s i t y of
C a l i forn i a .

S C HANE , Sanford A .
1972 ' Natural Rul e s i n Phono l ogy ' . In Robert P . S t o c kwe l l and
Ronald K . S . Macaulay , e d s , 1 9 7 2 , L � n g u����c C ha n g e a n d
G e n e�a � v e T h e o � q , 1 9 9 -2 2 9 . B lo omingt on : Indiana
Univer s i t y Pre s s .
 155

1973 G e n e�ati v e P h o n o l o g y . Englewood C l i f f s , N . J . : Pren t i c e -H a l l .

STANLEY , Richard
1967 ' Re dundancy Ru l e s i n Phon o l o gy ' . L a n g uag e 4 3 : 3 9 3 - 4 3 6 .

1973 ' Boundar i e s i n Phonol ogy ' . In Ander s o n and Kiparsky , e d s .

197 3 : 185-206 .

TEEUW , A .
1961 A C �itical S u� v e y 0 6 S tu d i eh o n Malay a n d B a hah a
I n d o n eh ia . Wi t h t he a s s i s t an c e o f H . W . Emanue l s .
( Konink l i j k Ins t i t uut voor Taa l - , Land - en Vo l kenkunde ,
Bib l i o grap h i c a l S e r i e s 5 . ) . The Hague : Mart inus N i j h o f f .

TRUDGILL , Peter
1975 Accent, Dialect a n d the S c h o o l . Londo n : Edward Arno ld .

WOLFF , John U .
1965 B eg i n n i n g I nd o n eh ia n . 2 vo l s . I t ha c a : C orne l l Univers i t y .
Lapoliwa, H. A generative approach to the phonology of Bahasa Indonesia.
D-34, vi + 160 pages. Pacific Linguistics, The Australian National University, 1981. DOI:10.15144/PL-D34.1
©1981 Pacific Linguistics and/or the author(s). Online edition licensed 2015 CC BY-SA 4.0, with permission of PL. A sealang.net/CRCL initiative.