^ h
3 1761041
c
j.
^1 cn^izzi^ ^^^H
1 -,
1 . '.
Hi
i •
1
i
■
!
y
MANUAL OF FRENCH
PRONUNCIATION AND DICTION
MANUAL OF
FRENCH PRONUNCIATION
AND DICTION
BASED ON THE NOTATION OF THE
ASSOCIATION PHONETIQUE INTERNATIONALE
BY
J. W. JACK M.A.
1^1 Xt^.
GEORGE G. HARRAP ?s' CO. LTD.
LONDON CALCUTTA SYDNEY
First published July ig22
by George G. Harrap 6r» Co., Ltd.
2 &*$ Portsmouth Street, Kingsway, London, W.C.2
ZIB7
A II rights reserved
PREFACE
THIS book is intended for pupils and students who
wish to acquire the normal or proper pronunciation
of French. Being an elementary work on scientific
lines, it offers a simple and reliable means of overcoming the
difficulties which the subject undoubtedly presents to English-
speaking people. It is a well-known fact that many pupils
who have an excellent knowledge of French grammar and
syntax, and have even passed the higher examinations, are
unable to pronounce French with any degree of success,
although the main purpose of learning a modern language
should be to speak it, and in such a way that the sounds may
be regarded as proper by a native. By a study of these
pages, a pupil can in due course reach as good results in
the way of pronunciation and accent as can be obtained by
a lengthy residence in France.
The pronunciation of French, like that of all other languages,
has no fixed or stereotyped form. It is exposed to the
caprices of fashion and taste, and varies in different districts
of France. The ' correct ' pronunciation is like the romantic
fairy, which vanished in smoke when one approached too
near. All competent judges, however, recognize that there
is a standard or normal French, viz. that of good Parisian
society or the educated classes in and around Paris and in
Northern France. This standard is for practical purposes
fairly uniform, and educated speakers nearly everywhere
tend to conform to it. It is this pronunciation, without
pedantry or vulgarity, that is described in these pages,
although it may not always be that given by certain dic-
tionaries. Even in the case of this standard, however, one
5
6 MANUAL OF FRENCH PRONUNCIATION
must not insist on dogmas ; for there are no two educated
persons in Paris who would be likely to agree in the pro-
nunciation of all French words. Many divergencies exist,
both in isolated words and in groups of words. In some
cases two, even three or four, forms of pronunciation of a
certain word must be admitted as equally good. But these
divergencies do not involve any fundamental difference in
the normal French referred to, and are noticed only by those
who care to watch for them.
The book is essentially a practical one. While it gives
the results of careful study made by distinguished French
orthoepists, it does not venture on any theoretical discussion
or scientific explanation of facts. The author has intention-
ally avoided all such subjects, which would only lead to
complication or confusion. As no exposition of pronuncia-
tion, however, can be sufficient without oral practice, he has
added suitable exercises and lists of words wherever necessary.
The use of the phonetic system of the Association Phonetique
Internationale, now recognized by the highest educational
authorities, including the English Board of Education, the
Scottish Education Department, and practically all collegiate
bodies both in Britain and the United States, will be found
of immense service. The science of phonetics, which has to
do with speech-sown^s, not only underlies all language, but
is a labour-saving device to the student, turning the com-
plicated difficulties of pronunciation into a simple system,
and enabling him to accomplish in a few weeks what would
otherwise require years. The written language gives only
a very imperfect and deceptive idea of the sounds, and the use
of reliable phonetics to the language-student is thus indis-
pensable, being a valuable aid in training both voice and ear.
The student should make a persevering practice of the
French sounds described in these pages. This is the secret
of proper pronunciation. To acquire only a knowledge of
the sounds is insufficient : they must be practised until their
correct use is second nature.
PREFACE 7
The author would take this occasion to record his indebted-
ness to the numerous masters of French diction from whose
works and teaching he has received suggestions and stimulus,
and particularly to Professor Paul Passy, Professor Maurice
Grammont, Professor Kr. Nyrop, Dr Ph. Martinon, L'Abbe
Rousselot, and other modern authorities, to say nothing of
Urbain Domergue, Mme Dupuis, Ch. Thurot, M. A. Lesaint,
and a galaxy of older writers. The Bibliography appended
may be of use to students who desire to extend their know-
ledge of French pronunciation.
Thanks are also due to Professor E. C. Hills, of the Uni-
versity of California, Professor Lander Macclintock, of Indiana
University, and Mr Alexander Green, for valuable suggestions
made on proof.
J. w. J.
CONTENTS
CHAPTER PAGE
Introductory: The Phonetic System . . ii
PARTI: THE VOWELS
I. The Fundamental Vowels . . . . i8
II. Close 7 [i] 26
III. Close E [e] 28
IV. Open E [e] . . . . . . . 30
V. Unstressed £....... 34
VI. Close A [a] 38
VII. Open ^ [a] . . . . . . . 42
VIII. Unstressed ^ ....... 47
IX. Close O [o] 49
X. Open O [o] 52
XI. Unstressed 0 ....... 55
XII. Close [u] 57
XIII. The Mixed Vowels : The Vowel [y] . . 59
XIV. The Vowel [0] . . . . . . . 62
XV. The Vowel [oe] 64
XVI. E Mute, or E Caduc . .... 66
XVII. The Semi-consonants : The Semi-consonant [w] 69
XVIII. The Semi-consonant [j] . . . . .71
XIX. The Semi-consonant [q] . . . . .74
XX. The Nasal Vowels ...... 76
XXI. Duration of Sounds, or Quantity ... 89
lo MANUAL OF FRENCH PRONUNCIATION
PART II : THE CONSONANTS
CHAPTER
XXII. Introductory .
XXIII. The Lateral Consonant L
XXIV. The Trills, [r] and [r]
XXV. The Fricatives .
XXVI. The Nasal Consonants
XXVII. The Explosives
XXVIII. The Letter H .
XXIX. The Numerals .
XXX. Double Consonants .
PAGE
99
103
107
III
121
126
138
141
144
PART III : WORDS IN COMBINATION
XXXI. Elision I49
XXXII. The Tonic or Rhythmic Accent . . .163
XXXIII. Liaison or Linking . . . . . .172
XXXIV. Emphatic or Supplementary Accent . . 184
XXXV. Assimilation . . . . . . .192
XXXVI. Intonation 198
XXXVII. Interrogations ...... 205
XXXVIII. Exclamations 209
XXXIX. Expression 212
XL. Rhythm and Eurhythmy . . . . .215
Index of Word-endings . . . . .223
Index of Principal Words Cited . . . 228
Index of Subjects ...... 234
Bibliography ....... 240
INTRODUCTORY
THE PHONETIC SYSTEM
1. The value of a phonetic system in the study of French
pronunciation is evident from the fact that the ordinary
^pelUng does not represent the actual sounds of the language.
No doubt, on the whole, French orthography is more con-
sistent in this respect than English, but it is by no means
adequate. There are only six characters {a, e, i, o, u, y) for
representing the vowels, but there are at least sixteen distinct
vowel-sounds in French ; and in the case of consonantal
characters, many of these are now extinct in pronunciation
or quite unstable. The word temps, for example, is written
with five letters, but in the actual spoken language it is
composed of only two sounds, t and nasal a [ta]. The only
letter in common is the t, for the spoken word contains neither
e nor m nor p nor s. Similarly the words eaux, oies, though
containing four letters each, are represented in speech by
the single sounds [o] [wa], which have nothing in common
with any of them. Many words too are spelt differently
but pronounced the same (e.g. sain, saint, sein, seing, ceins,
ceint, cinq) ; while others are spelt the same but pronounced
differently (e.g. Les fils ont casse les fils ; iu as /'as de
trefle ; les lacs sont pres des lacs).
2. All this confusion in the conventional method of spelling
French leads to difficulties in acquiring the pronunciation,
and hence we make use of a phonetic system as a guide to
the learner. The system is based on the principle of one
symbol, and only one, for each sound. Since every
sound in the language can be defined, it is evident that
we have only to allot a distinctive symbol to each one,
and we immediately possess an exact alphabet, more exact
indeed than the notation employed to represent musical
sounds. In this way we obtain a rational system of repre-
senting the spoken sounds, thus making them infinitely
11
12
MANUAL OF FRENCH PRONUNCIATION
easier to learn. In such a system, tan, tant, taon, tends,
tend, temps, being all pronounced the same, are indicated by
the same transcript [ta], while the various sound elements
of the language are all simply and accurately represented,
and the old difficulties of French pronunciation are largely
removed.
3. The alphabet used is that of the International Phonetic
Association, which is now the most widely used and highly
perfected of all Roman alphabets for indicating pronunciation.
It consists, for the purposes of the French language, of eighteen
consonants, three semi-consonants, and sixteen vowels.
Wherever possible, the same signs are used as in ordinary
spelling, and have practically the same value. This is the
case with the consonantal signs p, b, t, d, k, g, m, n, I, r,
f, V, s, z, and the aspirate h. But the following table
should be noted, containing three new consonantal signs,
and the three semi-consonants. These will be explained
later on.
Phonetic
sign
Ordinary
spelling
Examples
How pronounced
J
ch
chose, chaque,
cher
Like English sh {e.g. in shot).
5
j>g
jour, je, page
Like s in pleasure, z in
azure.
Ji
gn
ligne, regner,
Like ng in English sing, but
peigne
with point of tongue kept
down behind lower teeth,
and rest of tongue raised
swiftly against hard
palate.
J
y, i> 11
payer, bien, briller
Like English y in yet, yes.
w
ou, oi,
oui, fouet, moi,
Like English w in win,
[wa]
fois
dwindle.
q
u
lui, cuit, nuage
This is the compound vowel
[y ] (see table on p . 1 5) , but
sound so shortened as to
make one syllable with
the vowel immediately
.
following.
THE PHONETIC SYSTEM 13
On the principle of one symbol for each sound, it will
be understood that such symbols, as [k, g, s, z] are used
for various consonants or combinations in the ordinary
spelling. Thus :
[k] is used for ' hard ' c (i.e. c before a, 0, u, or another
consonant) ; for qu in such words as quatre, qualiU ; for
ch in such words as choeur, orchesire ; etc.
[g] is used for ' hard ' g (i.e. g before a, 0, u, or another
consonant) ; for gu in such words as guerre, langue,
Guillaume ; for c in second ; etc.
[s] is used for ' soft ' c (as in ici, lacet) ; for g (as in garfon,
franfais) ; etc.
[z] is used for s between two vowels (as in maison, causer) ;
for X in Haison (as in deux hpmmes) ; etc.
In such a way, by restricting the existing signs to one
sound each, and introducing a few new signs into the con-
ventional alphabet, it will be seen that the method of re-
presenting sounds is considerably simplified, and the study
of pronunciation facilitated.
4. The following table contains the phonetic signs repre-
senting the sixteen normal vowel-sounds of French. Of
these vowels eight are Fundamental, one {e mute) is
Indeterminate, three are Mixed, and four are Nasal.
All these sounds, like the consonantal ones, will be explained
in the following chapters, but the pupil is urged to acquaint
himself with them now as far as that may be possible, so as
to be able from the outset to read phonetic transcripts. The
column containing the ordinary spelling is not exhaustive,
but it will give some idea of the variety of method adopted in
current orthography to represent each sound.
The sign i after a vowel indicates that the sound is long.
Thus : [riiv], rive, [p£ir], pere. Sometimes it maj^ be necessary
in strict pronunciation to indicate half-length, in which case
one dot is used, as [vi'va], vivant.
Attention is called to the lip-rounding required in
some of the sounds, as indicated in the fourth column.
This is a distinctive characteristic of French, the lips being
much more used than in English, and it should be diligently
practised.
14 MANUAL OF FRENCH PRONUNCIATION
Fundamental Vowels
Phonetic
sign
Ordinary
spelling
Examples
How pronounced
i
i, y
Long : rive, pire,
green (long), react (short).
tige, bise
Not like i in bit, live, etc.
Short : vif, si,
hymne
e
6, e, ai, oe
Short: 6t6, les.
Scotch or Northern English
gai, oesophage'
a in take, case, etc.
American a in the first
syllable of aerial, aorta,
Shakespeare.
£
6, e, e, ai,
Long : pdre, tdte,
mer, faire, faite,
dairy (long), deaf {short).
ai, ay, ei,
ey
neige, Beyle
Short : d^, pret,
net, laid, plait,
rayon, peine,
Volney
a
a, k, a, e,
Long: page, part,
path (long), pat (short).
oi,oi[wa]
boire, cloitre
Scotch or Northern Eng-
Short: madame,
lish sound of these words.
1^, donndmes,
Also American sound of
femme, soif,
former.
cloitrer
a
a, a, oi, 01
Long : gaz, dge,
father (long). The short
[wa]
croire, croitre.
form is found in Southern
po61e
Short : pas, bat.
English diphthongs, as my
[mai], and also in Scotch,
trois, croit
Northern English, and
American sounds.
0
0, 6, au, u
Long : port, loge,
toge, Maure
shawl, tortoise (long) ; block.
waddle (short). English
Short : note.
and Scotch sound of these
hopital, Paul,
words. Also American
album
sound in door (long).
obey (short).
o
o, 6, au.
Long : chose.
Scotch or Northern English
eau
trone, cause.
0 in home (long), poetic
heaume
(short). Southern Eng-
Short : mot, deipot,
lish 0 in domain. Ameri-
haut, beau
can 0 in overawe, quarto.
u
ou, ou, ou
Long : rouge.
rude (long), July (short).
jour, douze
English and Scotch sound
Short : nous, vous.
of these words. In Am-
ou, gouter
erica the sound is found
short in whoever, rupee.
THE PHONETIC SYSTEM
Indeterminate Vowel
15
Phonetic
sign
Ordinary
spelling
Examples
How pronounced
9
e, ai, on
le, premier, fai-
sant, monsieur
Like e in over, taken ; or a
in about, again.
Mixed Vowe
LS
Phonetic
sign
Ordinary
spelling
Examples
How pronounced
y
u, eu
Long : sur, ruse,
Lips as for [u] {i.e. well
eurent
rounded and pushed out.
Short : plus, une.
as in whistling), but
eu, eus
sound the phonetic
vowel [i].
0
eu, eu,
Long : creuse.
Lips as for [0] {i.e. rounded
oeu
feutre, jeune
and pushed out), but
Short : deux, peu.
sound the phonetic vowel
voeu, jeuner
[e].
oe
eu, oe, oeu.
Long : fleur, ceil,
Lips as for [0] {i.e. slightly
ue
oeuvre, cueille
rounded and pushed out).
Short : jeune,
but sound the phonetic
oeillet, oeuvrer.
vowel [e]. When long, the
cueillir
sound approximates to
the Southern English one
in bird, turn, heard ; and
when short, to that in
love, glove, hurry.
For Nasal Vowels see next page.
5. French pronunciation, while based on the above sounds,
depends to some extent upon the division of words into
syUables. This division takes place according to the following
general rule : Finish with a vowel, and commence with
a consonant. Two consonants are divided, but if the
second be r or 1, both belong to the next syllable. Thus,
venir [v9-niir], cadeau [ka-do], sergent [ssr-3a], tableau [ta-blo],
nomhreux [no-br0], agneau [a-po]. In the case of two con-
secutive words, the same principle of division exists. Thus,
i6 MANUAL OF FRENCH PRONUNCIATION
Nasal Vowels
Phonetic
sign
Ordinary
spelling
Examples
How pronounced
a
an, am,
Long : tante,
Nasalized [a], i.e. with the
en, em,
jambe, lente,
soft palate hanging loose,
aon,aeii
temple
so as to allow some of
Short : blanc.
the sound into nose.
champ, cent,
temps, taon,
Caen
5
on, om,
Long : tondre,
Nasalized [o], but with a
un, um
nombre, punch
Short : bon, nom,
lumbago
little more lip-rounding.
I
in, im, yn,
Long : Inde,
Nasalized [ae] ([ae] is South-
ym, ain,
simple, lynx.
ern English a in such
aim, ein.
nymphe.
words as mad, rat, etc.).
eim,
vaincre, feinte,
(i)en,
Reims, pointe
oin
Short : fin, impur,
syntaxe, thym,
main, faim,
plein, bien, loin
&
un, um,
Low^ .-j'emprunte,
Nasalized [ae], with lips a
eun
humble
Short : un, par-
fum, k jeun
little rounded.
les enfants [le-zofS], une heure [y-noeir], while les aunes and
les zones are both [le-zoin].
Syllables may be either open or closed. A syllable is
said to be open when it is terminated in pronunciation by a
vowel. Such words as raison [re-zo], bonte [bo-te], vivacite
[vi-va-si-te], are therefore composed exclusively of open
syllables. A syllable is said to be closed when it is terminated
in pronunciation by a consonant, as in the words actif [ak-tif],
ternir [tsr-niir], percepteur [psr-SEp-toeir], where all the syllables
are of this nature. Open syllables are the more usual in
French, the tendency being to end a syllable on a vowel, as
forming the more audible sound. English, on the other
hand, has a preference for closed syllables.
THE PHONETIC SYSTEM 17
6. We give here a specimen of French in phonetic character,
followed by the same in common spelling. Although all
the sounds remain to be fully explained, the student should
endeavour meantime to understand them as far as he can,
and if possible to read the specimen with some degree of
accuracy, however small.
Phonetic Spelling
aprez avwar kree I9 sjel e la t£ir, dj0 a done la vi oz animo
e o plait ; pur le distege lez de dez oitr, il a vuly loer done de
no. oisi il loer a di d9 vniir de serte 3uir, a loer rakomadd
syrtu d9 n pa lublie. tuis 5 promi.
o 3ur di, le plait 5 defile dva Iqi, e elz 3 tut rasy ce no.
o dernje moma, kd tulmoid ete parti, ariiv a kura yn p9tit
plait do le petal ete d la kuloeir dy sjel. 3t0iz d9 sa foit, e1
S9 tne dvd so kreatoeir sdz oze Iqi dmdde koma si sapslre
dezorme, me dj0 s pdjd syr el kp a di dusmd, " n9 mubli
pa." la p9tit floeir e parti d kurd, e lorsk9 se kopap Iqi 3
dmdde so no, el loer a repody trist9md, " n9 mubli pa."
Common Spelling
Apr^s avoir cree le ciel et la terre, Dieu a donne la vie
aux animaux et aux plantes ; pour les distinguer les uns
des autres, il a voulu leur donner des noms. Aussi il leur
a dit de venir un certain jour, en leur recommandant surtout
de ne pas I'DubHer. Tous ont promis.
Au jour dit, les plantes ont defile devant lui, et elles ont
toutes regu un nom. Au dernier moment, quand tout le
monde etait parti, arrive en courant une petite plante
dent les petales etaient de la couleur du ciel. Honteuse
de sa faute, elle se tenait devant son createur sans oser
lui demander comment elle s'appellerait desormais, mais
Dieu se penchant sur elle lui a dit doucement, " Ne m'oublie
pas." La petite fleur est partie en courant, et lorsque
ses compagnes lui ont demande son nom, elle leur a r6pondu
tristement, " Ne m'oublie pas."
PART I: THE VOWELS
CHAPTER I
THE FUNDAMENTAL VOWELS
7. Vowels and Consonants. — It is very important to
understand the essential difference between vowels and con-
sonants. The student must not be led astray by the old
definition, which represents vowels as sounds which can be
uttered alone, and consonants as those which can only be pro-
duced in combination with vowels. The incorrectness of this
definition is evident, for there is nothing to prevent some
consonants {e.g. [s, /, f, v, 3], etc.) being uttered alone without
a vowel either before or after them. Unfortunately this
misleading definition is still widespread, and appears in the
word ' consonant,' i.e. ' sounding along with.*
A vowel [a, e, o, etc.] is the sound (proceeding from the
vocal chords) modified by the resonance chamber of the mouth,
there being at the same time a free passage for the breath.
On the other hand, a consonant [s, p, k, etc.] is a sound
caused by friction or stoppage of the breath somewhere in the
mouth. That is to say, in the former case the vibrating
breath is modified in the mouth, but not checked ; in the
latter case the breath is checked in various ways before it
leaves the opening of the mouth.
While the diiference between vowels and consonants is
thus distinct, there are some sounds which really belong to
both categories. In [m, n, 1, r], for example, while there is
a free passage of air, there is more or less a stoppage of the
breath (in [m] by closing the lips ; in [n] by raising the point
of the tongue ; and so on) ; but as the stoppage is the pre-
dominating quality, they are classified with the consonants,
making eighteen altogether. In the case of [w, j, q], there
is both a vowel and a distinct friction of the breath caused
18
THE VOWELS
19
by a narrowing of the passage, and hence they are classified
as semi-consonants (§ 3).
8. The Fundamental French Vowels. — Reserving the
consonants till afterwards, we now proceed to consider the
vowels, beginning with the fundamental ones. These, as
already stated, are eight in number : [i, e, e, a, a, 0, o, u].
We give a general description of them in this chapter, followed
by a detailed consideration of each one in the immediately
ensuing chapters. In order to describe them clearly and
accurately, we print in each case a front view of the lips
and a side view of the mouth, along with French words in
phonetic notation containing the sounds. The student
should set himself to acquire correctly the distinctive pro-
nunciation of each of these fundamental vowels. Each
one should be dwelt on decisively, to make sure that the
true sound has been caught, and it must be repeated till
familiarity is acquired.
English green (long) , react (short)
Phonetic : [di, li, si, sim, mil, fis, midi, kritik, mirifik,
riiv, tii3, liir, biiz, 3it, fiij, biij, filip].
Same in ordinary spelling : dit, lit, si, cime, mille, fisse,
midi, critique, mirifique, rive, tige, lire, bise, gite, fille,
bille, Philippe.
Scotch or Northern English a in take
Phonetic : [de, Je, 36, pre, kle, ete, prefere, repete, desede,
egliiz, Use, felisite, efemine, grije, milje].
Ordinary: des, chez, j'ai, pre, clef, et6, pref6r6, repete,
dec^d6, eglise, lycee, felicite, effemine, griller, miUier.
20 MANUAL OF FRENCH PRONUNCIATION
English dairy (long), deaf (short)
Phonetic: [set, Jef, rsspe, etsrnsl, meir, bei/, tsit, teir,
tel, bek, net, perpleks, fidel, tristes, regie, pest].
Ordinary : cette, chef, respect, eternel, mer, beche, tete,
terre, tel, bee, net, perplexe, fiddle, tristesse, regie, peste.
a
Path (long), pat (short), Scotch or Northern English
sound of these words
Phonetic : [sal, bat, nap, madam, kanal, kamarad, malad,
pai3, kaiv, ami, ale, imai3, asje, laskair, aspe].
Ordinary : salle, batte, nappe, madame, canal, camarade,
malade, page, cave, ami, allee, image, assied, lascar, aspect.
'"fi
a
English father (long) ; no English case of short,
except in diphthongs
Phonetic: [pa, ka, 3ain, faiz, gaiz, baiz, aipr, gate,
dega, gate, klima, ama, 3adis, pulaje, espais, maren].
Ordinary : pas, cas, Jeanne, phase, gaz, base, apre,
gater, degat, gateau, climat, amas, jadis, poulailler, espace,
marraine.
THE VOWELS
21
3
English shawl (long), block (short)
Phonetic : [od, rob, rof, fDrin, bi5, foir, moir, oktobr,
etof, epok, parte, modest, poeim, oval, kuron, omon].
Ordinary : ode, robe, roche, forme, loge, fort, mort,
octobre, etoffe, epoque, porte, modeste, poeme, oval,
couronne, aumone.
Scotch or Northern English o in home (long),
poetic (short)
Phonetic : [do, mo, to, foljo, /oiz, roiz, doim, trom, eko,
zero, nivo, sito, kuto, trsno, bsstjo, Japo, alto].
Ordinary : dos, mot, tot, folio, chose, rose, dome, trone,
echo, zero, niveau, sitot, couteau, traineau, bestiaux,
chapeau, alto.
U
English rule (long), good (short)
Phonetic : [ru, mu, su, rut, tut, pul, rukul, kurt, mus,
buk, sup, rui3, kuir, duiz, rutin, dute, tupE, rulad.]
Ordinary : roue, mou, sou, route, toute, poule, roucoule,
courte, mousse, bouc, soupe, rouge, cour, douze, routine,
douter, toupet, roulade.
9. Note that these vowels have been arranged according to
the position of the tongue, mouth, and lips.
22 MANUAL OF FRENCH PRONUNCIATION
(i) The Tongue. — For [i] it will be noticed that the tongue
is raised very high in front. It is lower for [e], lower still for
[s], and lowest of all for [a] and [a], where it approaches the
flat position as much as possible. It now begins to rise
again for the remaining vowels (this time at the hack), being
higher for [o], higher still for [o], and highest of all for [u].
In pronouncing the eight vowels in the order mentioned, one
can easily verify these tongue-movements, especially with
the help of the finger or a pencil.
The vowels [i, e, e] are consequently called front vowels,
and [d, o, u] back vowels.
(2) The Mouth. — For [i], it will be observed, the mouth is
somewhat close. It opens more for [e], more still for [e],
and most of all for [a] and [a]. It then begins to close again,
a little for [o], more still for [o], and most for [u]. The open-
ing or closing depends on the movement of the lower jaw,
and may easily be noticed with a mirror.
Hence we distinguish two classes of vowels, close and open.
It is evident that in general those at the top and bottom of the
series are close, while those about the middle are open ; but
looked at from another point of view, the vowels maybe arranged
in pairs. Thus [e] is close e, while [e] is open e ; [o] is close 0,
while [o] is open 0 ; [a] is close a, and [a] (which has the
mouth widest of all) is open a. The vowels i and u are only
found close in French, as a rule, but the open sounds of these
occur frequently in English (e.g. i in hit, hinge ; u in pull, full).
This distinction between close and open is an essential
one in French, and should be clearly understood. To give, for
example, a close e for an open one (in such words as pere, tele,
etc.) is common with beginners, especially in districts where
close vowels predominate, but is incorrect.
(3) The Lips. — The shape of the lips in the series is very
important. For the front vowels, it will be noticed, the lips
are somewhat fiat, their ends being drawn back, while for
the back vowels the lips are distinctly rounded and pushed
out — a little for [o], more still for [o], and most for [u].
For the front vowels in this series there must on no account
be any rounding and projection of the lips, as this would
produce vowels of a different kind, namely compound ones
(§ 58), but in the case of the back vowels this particular quality
is absolutely necessary. As it is less frequently found in
English, the student should practise it well. Special oppor-
THE VOWELS 23
tunities will be given for this later on, when we come to treat
individucdly of the back vowels.
10. For the reasons just enumerated, the French vowels
may be disposed in the form of a triangle, which shows more
clearly the relative positions of the tongue :
Front Vowels^ " I/^^^^-^^^arA: Vouds
"^036)1
'(a) (Close,)
(Open) CS. hu)iOpe.n)
(Close) e\. ko)(Closz)
p)(Opcn)
The curved line represents the roof of the mouth. The centre
at the bottom (at [a]) represents a neutral position, where the
tongue is lowest and the mouth widest. Beginning from [a]
and going up to the left indicates a gradual rising of the tongue
more and more towards the front (the hard palate), with a
corresponding closing of the mouth, while going up to the
right means a gradual rising of the tongue more and more
towards the back (the soft palate), with a corresponding
closing of the mouth by rounding and projecting the lips.
This last characteristic is indicated by curved brackets ( ).
11. Tenseness. — ^The nature of all vowels is considerably
affected by the condition of the tongue muscles. If these
muscles are braced up and held firmly in position, the vowel
is said to be tense. If they are loose and relaxed, the vowel
is called slack.^ A tense vowel has a clear, firm sound, and
a slack one a dull, weak sound. It is possible to pronounce
all vowels either way.^ In French, however, practically all
the vowels are tense, in the sense that they are pronounced
clearly and with a distinct muscular tension. This is par-
^ The terms narrow and wide are used by some phoneticians instead.
* In accurate phonetic script the grave accent is used to denote
slackness. Thus [h] is slack [e].
24 MANUAL OF FRENCH PRONUNCIATION
ticularly so with the close vowels, the closeness being partly-
due to the tension. In this respect the French vowels differ
from many of the English vowels, especially the short ones,
which are produced generally with the tongue relaxed. In
French there are exceptions, and these will be referred to later
on, but the student should understand that the funda-
mental vowels in the preceding series are all pronounced tense,
the close ones especially so, the open ones not so much.
12. Uniformity. — Many vowels in English, especially long
ones {e.g. in such words as case, note, day, etc.) are pronounced
as if they were diphthongs — i.e. they become in reality a
combination of two sounds in one syllable. This is not so
common, e.g., in Scotland, where the vocal organs are held
more tensely and do not glide during the emission of a vowel,
but in England and America the diphthongal tendency is very
prevalent. It cannot, however, be too strongly enforced that
in French all vowels, whether long or short, are uniform
throughout their utterance. There are no diphthongs in
French, except in rapid conversation, and even then in quite
exceptional cases. If two vowels are in contact, they are
pronounced as two distinct syllables, e.g. poete [post], hahut
[bay], plier [plie], or else one of them becomes a semi-con-
sonant, e.g. souhaii [swe], rien [rje]. French vowels are
practically all simple, uniform sounds, the organs of
speech being kept quite still in their original position
during the production of them. They must be pronounced
exactly the same from beginning to end, without being allowed
to glide into another sound. Unless the beginner is exceed-
ingly careful, he will find himself at fault here. He will
probably produce diphthongs * a Tanglaise ' without being
aware of it. It is common, for instance, to hear a beginner
pronouncing mere as [msia] or [mear], instead of [meir] with
an unchanging vowel sound.
13. Clearness. — ^The French language is distinguished from
most others by the clearness or crispness of its vowels, which
stand weU out from their surroundings. Only one vowel,
the mute e, is vague and tends to disappear. AH the others,
even though uttered lightly and briefly, must be given with
definiteness and distinctness, more so than in English. The
advice, " Look after the consonants ; the vowels will look
after themselves," is ruinous in French, where every vowel-
THE VOWELS 25
sound is clear and definite, and none is the least muffled
or less audible than the others. French consonants may
change and even disappear from pronunciation (as in many
finals), but the vowels, whether long or short, stressed or un-
stressed, preserve their distinctness. This is a rule to which
the learner should pay special attention. No vowel should
be slurred over, or lose its clear quality. A crisp enunciation
of the vowels alone will make anyone's French remarkably
good.
14. Sufficient has been said in this chapter to show that
the organic basis or general characteristic of French pro-
nunciation is different in many ways from that of English.
In French the vocal mechanism has its own peculiar action
and tendency, producing in some cases results rarely, if ever,
found in English. The lips, as we have seen, are very active,
sometimes contracting and protruding quickly and energetic-
ally. The vowels are generally tense, not lax as in English,
and are clear uniform sounds, not diphthongs. It is evident
that a true pronunciation of French cannot be acquired without
persevering practice of these special movements. The vocal
muscles require to be trained and developed on the French
basis. This cannot be done in two or three weeks : it requires
a long course of careful practice. It is here that one great
advantage is found in the phonetic system, which acts as a
valuable aid in such training, and makes up largely for the
want of residence in France.
CHAPTER II
CLOSE I [i]
15. We now take each vowel separately and in detail,
beginning with [i], at the top left corner of the triangle (§ lo),
and ending with [u]. The French [i] (represented in ordinary
spelHng by i, i, i, y) is close (§ 9 (2)), being identical with
the English i in marine, police, clique, fatigue ; ee in eel, deep,
seen ; ea in head, leak, leave. It may be long, as in the English
words just quoted, or short, like e in helow, delay, return.
Open i, on the other hand, is not the short of the close one,
as some beginners imagine : it is quite a different sound,
found in such English words as ill, dip, sin, hid, lick, live.
It is formed with the mouth more open, and the tongue a
little lower in front. The two sounds have little in common
except the orthographic symbol.
For the French [i], the tongue (see diagram, § 8) is pressed
strongly against the lower incisive teetH, while the front part
* bunches up ' and presses against the upper molars and the front
part of the hard pcdate, leaving a very narrow passage at the top
for the air current. On this account it is known in phonetic
language as a ' high-front vowel.' The mouth is not much
open, being flat, and the corners of the lips being drawn back.
The muscles are held more or less tense (§ 11).
The chief fault of the beginner lies in substituting the
open sound for the close one. The former rarely if ever
occurs in French, and besides being strange to a Frenchman, is
often mistaken by him for the close [e] or the compound
vowel [y]. Another fault is to forget the tenseness and pro-
nounce the sound lax as it is in English.
16. When [i] is in the last syllable of a stress group,
i.e. when it receives the Tonic or Rhythmic Accent,^ it is
1 In French, unlike English, all syllables are pronounced with almost
equal emphasis, so much so that the language might be called
'* monotonous," but at the same time a very slight stress falls upon
26
THE VOWELS
27
always long if followed by [v, z, 3, j], or final [r]. In this
respect it follows the seven-vowel rule [i, 8, a, o, u, y,oe]
(§ 105 (4)). Thus, gencive [3asiiv], grise [griiz], vertige [ver-
tii3], hille [biij], dire [diir]. Many people pronounce it long
also in those words in -is (mostly classical) where the s is
sounded, as Daphnis [daf niis] , gratis [gratiis] . In all other cases
it is short, even though circumflexed, as lit, ami, hrique, myrte,
asile, gUe, Ntmes, ftmes, fttes. An exception is abime [abiim].
When unstressed, it is generally short and pronounced
with less muscular tension, frequently becoming half-open.
Examples : ici, fini, pigeon, midi, Rivoli, mi-partie, it y va,
diligence, etc.
EXERCISES
i
Je dis, tu dis, il dit, nous
disons, vous dites, ils
disent
39 di, ty di, il di, nu dizo, vu
dit, il di:z
Cire, mise, corrige, livre,
anguille
si:r, mi:z, kori:3, li:vr, agi:j
Baucis, Paris, Davis, Osiris,
bis, lis, iris, oasis, metis
bosiis, pari:s, daviis, oziriis, bi:s,
li:s, iri:s, oaziis, metiis
Digne, guide, miche, lisse,
libra
diji, gid, mij, lis, libr
1
Naif, hair, Jamaique, Moise,
Sinai
naif, ai:r, 3amaik, moi:z, sinai
i
Dime, dine, ile, epitre
dim, din, il, epitr
y
Cycle, myrte, crypte, lyre,
analyse, lycee, synonyme,
thyrse, scythe
sikl, mirt, kript, li:r, anali:z,
Use, sinonim, tirs, sit
11 lit le livre parmi les myrtes.
La fille de Cyrille a six prix.
11 vit I'eclipse d'ici en avril.
Le guide dit que le tigre est assis.
Voici les archives de la ville.
il li l9 li:vr parmi le mirt.
la fi:j da siril a si pri.
il vi le klips disi an avril.
l9 gid di k9 la tigr et asi.
vwasi lez arjiiv da la vil.
the last syllable of a word, or group of words, containing a single
idea. This stress is termed the Tonic or Rhythmic Accent (accent
d'intensite) , and the word or group of words is known as a stress-
group. All syllables except the last one are said to be unstressed.
CHAPTER III
CLOSE E [e]
17. The vowel e in ordinary spelling frequently represents
the so-called e mute, which is often not sounded, as pensera
[pasara] or [pasra]. Apart, however, from this indeterminate
form, the vowel e comprises at least two distinct classes —
close, as in ete, and open, as in pere.
18. Close [e] (represented in current spelling by e, e, ai, cb)
is similar in pronunciation to the Northern English or Scotch
a in such words as case, ache, bathe, take, etc., but pronounced
short. In the south of England and other parts of the
English-speaking world these words, besides being sounded
open, are turned into a diphthong ([£]-f [i]), which is foreign
to French, but the northern sound is close, monophthongal,
and practically identical with the French one.
19. Referring to the diagram in § 8, it will be seen that the
tongue is a trifle lower than for [i], but its tip still presses
against the lower incisive teeth and its front part against the
upper molars. The mouth is a little more open, and the lips
are not stretched so flatly. The muscles of the tongue and
mouth are held tense, and it is here specially that English
students are at fault. Not being accustomed to a tense [e],
they allow the sound either to glide into a diphthong or to
approach loosely to the open e (the sound in English dairy).
The correct pronunciation of close [e] demands an effort of
the vocal muscles if this fault is to be avoided.
20. In correct French, close [e] occurs only in open
syllables, and is nearly always pronounced short. It is only
when special emphasis is laid upon the syllable containing it
that it may become long, as in dejd I [dei3a], gaiete [geite],
gueri ! [geiri].
21. It is generally found as follows in the ordinary spelling :
THE VOWELS
29
' (i) As (? with the acute accent : Thus, ble [ble], dS [de,] bibi
[bebe].
(2) As e followed by a final mute consonant other than / :
Thus, clef [kle], assez [ase], porter [porte], pied [pje],
vous parlez [vu parle], je m'assieds [39 masje]. Also in the
conjunction et.
In ordinary familiar speech, the monosyllables les,
des, mes, ies, ces, ses conform to this rule, being pro-
nounced close ; but in elevated or emphatic speech the
open sound [e] predominates, and generally becomes
long [ei] on the stage.
(3) As ai in certain verbal terminations (futures and the past
definites of first conjugation) : Thus, faurai [3ore],
fallat [3ale], je parlai [parle], je niendrai [vjedre], j^ai
[3e]. Also (in Paris, at least), in gai, gate, geai, je sais,
tu sais, il sail ; and frequently in quai, pays, abbaye
[pei, abeji], je fais, tu fais, il fait, je vais?-
(4) As Greek oe : Thus, cecuminique [ekymenik], cedeme
[edeim], cesophage [ezofai3].
(5) As final in -oe, as Crusoe [kryzoe], FSroe [feroe].
^ The tendency, however, in all these cases nowadays, except in j'ai,
is to open the vowel.
EXERCISES
e
The, fee, aime, annee, fumee,
seme, pre, donne, cliante,
bonte, repete
te, fe, sme, ane, fyme, same, pre,
done, fate, bote, repete
e
Duprez, chez, nez, vous
avez, rez
Berger, dernier, clocher,
volontiers, rosier, parler,
Roger, Alger, Tanger
Assieds-toi, tu t'assieds, il
s'assied
dypre, Je, ne, vuz ave, re
b£r3e, dernje, kloje, volotje,
rozje, 'parle, ro3e, al3e, tase
asje-twa, ty tasje, il sasje
ai
Je portai, je porterai, je
jouai, je prendrai, je men-
tirai, je serai
39 porte, 39 portre, 39 3we, 39
pradre, 39 matire, 39 sare
Je donnai les clefs au bebe.
11 a seme le ble dans le pre.
J'ai assez de the pour les fees.
Je chanterai chez vous en ete.
39 done le kle 0 bebe.
il a same l9 ble da I9 pre.
3e ase d9 te pur le fe.
39 Jatre Je vu an ete.
CHAPTER IV
OPEN E [8]
22. This is a vowel-sound heard both in the south of Britain
and in America. It is found long in such words as there, pair,
hear, fairy, dairy, and short in deaf, dead, said, tread, etc. It
is the vowel of the sheep's bleat. For correct articulation the
mouth requires to be more open than for close [e]. In other
words, it is necessary to drop the lower jaw a little. The tip
of the tongue no longer presses against the lower incisive
teeth, but against their base, while the front part does not
rise so high towards the palate. The tenseness is not so
great, the tongue muscles being fairly relaxed, and no special
effort being needed.
23. In the enunciation of the long open e most EngUsh-
speaking students are at fault. They generally pronounce
it as a diphthong, allowing the sound to modify itself
during its emission, so that in place of [teit] {tete) one hears
something like [teit], or [tait]. For the correction of this
fault it is only necessary to watch that the tongue remains
exactly in the same position during the duration of the sound
(§ 12). In some parts again, such as Scotland, there is really
no long open e, and the vowel is turned into the close one, so
that mere [meir], pere [psir] become [mer], [per]. This fault
can be largely remedied by exercising less tension in the
tongue muscles.
24. Open e is represented in the ordinary French spelling
by a great variety of symbols : e, e, e, ei, ai, at, aie, ay, aye,
ey. In regard to length, it follows the seven-vowel rule
(§ 105 (4)) — that is, when it is in the last syllable of a stress-
group, it is always long if followed by [v, z, 3, j] or final [x]}
In addition, it is generally long :
^ When it is followed by ;', students should guard against the mistake
of pronouncing it like e in the English word certain. It should rather
30
THE VOWELS 31
(i) In the termination -es (mostly foreign words), where the s
is sounded, as Pirides [perikkis] Ramses [ramseis], ftores
[floreis], pataques [patakeis].
(2) In closed circumflexed syllables, as reve [rsiv], fete [feit],
enchaine [qfem].
(3) In the terminations -aisse, -erne, -ene, and sometimes in -eine,
-aine. Thus, haisse [beis], creme [kreim], scene [sem], reine
[rem], gaine [gem].
When unstressed, it is usually short, whether circumflexed
or not, as rester [rsste], faitage [fEtai3], aigreur [egroeir], frai-
cheur [fre/oeir], le meme homme [b mem om]. When short, it
is hardly so open as when long, unless followed by two
(sounded) consonants. But these rules must be regarded as
merely general, as open e has many variations, for which the
student is referred to the chapter on Duration.
25. As just stated, many symbols are used in ordinary spell-
ing to represent the open e. The following always denote it :
(i) The letters e, e, e (except as in § 21 (5)), and the groups
ei, ai, as regie [regl], exces [ekse], pr ogres [progre], chene
[Jem], Noel [noel], neige [nei3], maiire [meitr].
(2) The letter e :
In closed syllables, as tel [tel], bref [bref], hec [bee].
Before a * double consonant,' as gemmer [3eme], ierreur
[terceir].
In the terminations -et and -ect, as projet [pro3e],
aspect [aspe], direct [direkt]. Also in il est [e],
tu es [e].
(3) Close e falling in a closed syllable, as parli-je [parlei3],
dussi-je [dyseij], aimai-je [emei3], dirai-je [direi3], ai-je
[ei3], que sais-je [k9 sei3]. The change from the close to
the open is due to the fact that the French have a dis-
inclination to the close sound in a shut syllable. It is
for the same reason that 6 in verbs becomes e when the
syllable closes, as esp^rer, but fespere.
(4) The group ai, except in the few cases where it is
close (see preceding chapter), as vrai [vre], paix [pe],
laide [led], je chantais [39 Jate].
be like the sound of e in herring. Thus fermer is [ferme], not [foerme].
Practice in this matter should be made with such words as exercer^
exercice, merer edi, conversation ^ per mis, persuader, universel, etc.
32 MANUAL OF FRENCH PRONUNCIATION
(5) The groups ay, ey. Before a consonant or in a final
syllable they are pronounced [e], as Raymond [remo],
Leyde [led], Fontenay [fotne], bey [bs]. Before an audible
vowel they become [ej],^ as rayon [rsjo], frayeur [frejoeir],
seyant [seja]. Before e mute they are sounded [sij],^ as la
paye [lapEij],z7 raye\}\xz\]\,ils grasseyeni\\\gr2isz\]]. But in
La Haye and a few other proper names they are sounded
[e] as [la e], etc.
(6) The group aie, as haie [be], fuiaie [fyte], Us voudraieni
[il vudre]. When, however, aie represents -aye (from
verbs in -ayer), there is a tendency to pronounce it
[ej] (or [eij] when final) in accordance with preceding
rule, as il paie [il peij], Us essaieni [ilz eseij], higaiement
[begejma]. Similarly, in the verbal form eie, still occasion-
ally seen (from -eyer verbs), the sound [eij] is common,
as fasseie [3aseij], Us asseient [ilz aseij].
26. When open e is final, many French speakers prefer to
sound it close, as mais [me], jamais [3ame], succes [sykse],
stijet [sy3e], etc. In the finals of conditionals, however, open c
is maintained, to avoid confusion with futures in -at.
* Except in bayadSre, mayonnaise, fayard, bayer aux corneilles, and
some proper names such as Bayard, Bayeux, Bayonne, Cayenne, etc,
and thsir derivatives, where a>' = [aj].
* Except in cipaye [sipa:j], cobaye [koba:j], Biscay e [biska:j], and
one or two other proper names .
THE VOWELS
33
EXERCISES
h
e
ei
ai
Tr6ve, college, amere, fiddle
Beche, fenetre, foret, etre, bete
Israel, Joel, Laerte, Raphael
Seigle, haleine, peine, beignet
Faite, paitre, nait, parait
treiv, kol£:3, ameir, fidel
be:]", faneitr, fore, e:tr, be:t
israel, 308l, laert, rafael
segl, alen, psn, bejie
feit, pEitr, ns, pare
e
Sec, spectre, dette, bel, berge,
serpe, ellipse, ferr6, message,
sellier
Crochet, collet, bonnet, chevet,
mets, regret, secret, respect,
jet, cadet
sek, spektr, det, bsl, ber3,
SErp, slips, fere, m8sa:3,
selje
kroje, kole, bone, Jave, me,
ragre, sakre, respe, 3e,
kade
ai
Aimer, air, balai, palais, je serais
eme, eir, bale, pale, J9 sare
6
Donne-je, serai-je, repondrai-je
done:3, sare:3, rep5dr8:3
ay
ey
fipernay, Launay, balayer, be-
gayer, il effraye, il 6taye, ils
bdgayent
Volney, jockey, ils seyent,
asseyez, trolley, Jersey, Bom-
bay, Bellay
eperne, lone, baleje, begeje,
il efreij, il eteij, il begeij
volns, 3oke, il sejj, aseje,
trole, 3erz8, bobe, bele
aie
Claie, haie, laie, plaie, craie, taie,
ils auraient, ils mangeaient
kle, e, le, pie, kre, te, ilz ore,
il ma38
Elle aime son pere, sa m^re, et ses
fibres.
Elle parait etre une belle reine.
Quel est le libraire de Geneve ?
Faites finir le bonnet et le gilet.
Le merle chantait hier dans la foret.
el e:m s5 pe:r, sa me:r, e se
freir.
el paret e:tr yn bel rem.
kel E la libreir da 3an8:v ?
fet fini:r la bone e la 3ile.
la merl JatE je:r da la fore.
CHAPTER V
UNSTRESSED E
27. Normally, the vowel e, when unstressed, maintains its
close or open sound as the case may be, although the close
sound is pronounced with rather less tension. This is
specially the case in many isolated and independent words.
Thus, it is close in egal, geant, fletrir, desir, Staler, defense, etc. ;
and open in personne, merci, seigneur, domestique, Neptune,
etc. The same is true of derived words, which follow the stem
in cases where they are not subject to any modifying influence.
Thus, the vowel is close in gueable (from gue), gaiete or gaUe
(from gai), feerique (from fee) ; and open in fratcheur (from
frais), aigreur (from aigre), faitage {hov[ifaite).'\
28. But very often close e or open e in unstressed syllables is
replaced by a sound intermediate between the two. For
instance, in reveler, only the final e is really close, and in pieton,
messieurs, etc., the e is not quite close either. Similarly, many
vowels that one might expect to be open become half -close.
There is a distinct difference, for example, between the vowel
[e] in the infinitive aimer and that in faime. Not only is the
former shorter, but also less open than the latter. The same
may be said of trailer and traitent. The sound referred to is
known as middle e. It is less tense than the close [e], being
pronounced with mouth a little more open and the front of
tongue a little lower, but not so much as to produce [e]. The
sound may be placed on the triangle (§ lo) midway between
[e] and [&], although in some cases it seems to approach nearer
to [e] and in others to [e], being sometimes confused with one
or the other. In England it resembles the vowel in the
words men, net, ferry (first syllable), though American speech
makes it open in such cases. In French it occurs only
in unstressed syllables, where it is very frequent. The
phonetic symbol for it is [e] ; but after all, the difference in
sound is so slight that the ordinary symbol [s] may generally
34
THE VOWELS
35
be used instead, and the difference ignored, especially by
beginners.
29. One of the principal causes leading to middle e is vocalic
assimilation. The vowels contained in two consecutive
syllables have a tendency to assimilate in timbre, the first
taking almost the same timbre as the second. Thus in traitent
the vowel is open according to the general rule, but in traiter
it assimilates itself somewhat to the second vowel which is
close, thus becoming only half-open. The following table
contains examples of this tendency. It will be observed that
the change is mostly from the open sound towards the close
one, though cases of the opposite are not infrequent.
Open
Open
Less open
Open
Open
Less open
Bete
betement
betise
Fele
felons
felure
Presse
pressons
presser, -ez
Guet
guetteur
guetter, -ez
Fete
fetons
feter, -ez
Laisse
laissons
laisser, -ez
Peche
pecheur
pecher, -ez
Baise
baisons
baiser, -ez
Greffe
greffoir
greffer, -ez
Baigne
baignoire
baigner, -ez
Quete
queteur
queter, -ez
Plaire
plaisant
plaisir
Pret
pretons
preter, -ez
Paix
paisible
Aise
aisance
aise
Tete
teti^re
tetu
Grele
grelon
greler, -e
Aigle
aiglon
aigu, aiguille
Aile
aileron
aile
Mais
••
mais oui
Close
Close
Less close
Close
Close
Less close
ete
etiez
etais [ste]
ebeniste
ebenier
ebene [ebem]
elever
elevez, -e
el eve [eleiv]
begayer
begayez, -e
b6gaie [bege]
30. Apart from vocalic assimilation, the vowel e in ordinary
spelling when unstressed very frequently represents a
middle e, or a sound so open that it may not incorrectly
be written as [s].^ This is so in the following and in other
cases :
(i) Before r followed by another vowel, as jeroce [feros],
exterieur [eksterjoeir], preference [prefsrais], espirance
[esperais], mysiirieux [mistErj0], sirieux [s£rj0], vereux
[v£r0]. In many cases, however, there is hesitation.
^ Grammarians generally represent e (with an acute accent) as always
close, whether stressed or not, but as a fact it is only close for certain
when final.
36 MANUAL OF FRENCH PRONUNCIATION
(2) Before a consonant followed by a silent e mute, as
midecin [metse], hSqueter [bskte], Svinement [evenmcl],
vSnerie [venri], dleri [selri], imeri [emri]. Properly
speaking, such words should have the grave accent owing
to the syllable being practically closed (see § 25 (3)),
but the grave accent is rarely written in the middle of
a word.^
There are however exceptions, as ilever [elve], dimener
[demne], Smeraude [emroid], ichelon [eJ15], crineler
[krenle].
(3) When it represents e in the root word or stem. Thus,
we have rSgler [regie] from regie, rigner [rejie] from regne,
sScher [seje] from seche. Similarly, we have mitrer,
dder, icritner, inquUter, ISguer, etc., pronounced with
fairly open e.
When initial, however, i is often less open than when
in the interior of a word.
31. In many cases the unstressed e is variable, some
speakers pronouncing it open, others intermediate, and others
close in the same word. Thus :
(i) The symbol at is generally open, as maison [mszo], raison
[rezo], comhinaison [kobinezo], raisin [reze]. But in a
few words of this category the middle or close e is also
common.
(2) The same is the case with the combination es, which varies
between the open sound and the close one, as esprit
[espri], [^spri], or [espri]. Thus also with such words as
estimer, essentiel, essayer, respect, rester, restreindre, res-
pirer, resplendir, ressusciter, desservir, dessicher, dessiner,
descendre, pressentir, messeoir, message, domestique,
forestier, Robespierre, etc.^
(3) The prefix ex- before a vowel or h ' mute,' and the prefix
ef- before / are often pronounced with the sound inter-
mediate or close, as exercice [egzersis], examiner [egzamine],
1 In the Dictionnaire de V Academic (1877), avSnement was altered to
avinement, but many other words, such as ivinement, were left un-
altered.
2 It should be remembered that there are numerous cases of es where
the vowel represents e mute, as in dessous, dessus, ressembler, ressort,
etc. These are the prefixes de- and re- before s, which is doubled to
preserve the s sound (§71 (S))*
THE VOWELS 37
e§et [efe], effort [efoir], inexact [inegzakt], inexigihle
[inegzi3ibl], inefficace [inefikas].
32. It will be seen from this chapter and the two preceding
ones that the vowel e varies greatly between the very open
sound and the very close one. In particular, unstressed e
has numerous individual divergencies and many delicate
shades. But the beginner need not trouble too much over
these. He cannot go far wrong though he disregards vocalic
assimilation, variability, and similar matters, and includes
under [e] all cases referred to in this chapter. He will not
commit a grave fault if he opens the vowel a little more or a
little less.
33. Unstressed e frequently becomes mute in rapid, careless
pronunciation. This is particularly the case in monosyllables,
such as Us, des, cet, cette, est, etc., and even in such a word as
dejd. In popular print one finds such forms as I's hommes
[Izom], d's auires [dzoitr], cet enfant [stdfa], cette histoire
[stistwair], c'est-d-dire [stadiir], il est dejd venu [il 8 d3a vny],
etc. Needless to say, this is a freedom of language that
should be avoided.
On the other hand there are several words in which an
apparent e mute is sounded half -open e : huffleterie [byfletri],
marqueterie [marketri], parqueterie [parketri], mousqueterie
[musketri]. Properly speaking, these words should have
double t. Similarly, we have senegon [senso], chevecier [/evsjc],
Genevois Ijenvwa], etc.
CHAPTER VI
CLOSE A [a]
34. There are two a sounds, close and open, as there are
two kinds of the other vowels. Close a [a], is the Scottish or
Northern English pronunciation of the vowel in such words
as act, bat, pat, mat, where it is short, and hath, past, mast,
where it is long. In America the short sound is not usual,
[ae] being substituted for it as in the South of England, but
the long sound is heard in such words as those just mentioned
(hath, past, etc.). The open a [a] is quite a different sound,
much broader, found in the English words father, psalm,
calm, etc. The difference between the two sounds is much
less apparent to the ear than between the two e's, with the
result that many people confuse the two.
35. For [a] the mouth is not so open, nor the tongue quite
so low down as for [a], which is the widest of all vowels and
has the tongue lying as low as possible. In the case of [a]
the lips are rather flat, with their corners slightly drawn back,
while for [a] the lips become rather rounded, with their corners
drawn in. For [a] again, the tongue muscles are held rather
tense, with the point of the tongue slightly advanced and
sensibly pressed against the lower incisive teeth, while for [a]
there is only sufficient tenseness to maintain the sound.
36. It requires to be emphasized that the close [a] re-
ferred to is that found in Scotland or Northern Eng-
land. The tendency of all Southern English and American
English is to substitute [se] for the short sound. The [se] is
intermediate between [e] and [a], and does not occur in
French, unless as a nasal (§ 97). The English speaker, for
example, pronounces the words pat, rat, as [paet], [raet]. This
substitution of [se] is a fault to be avoided in French. The
close [a] is so frequent in French that any negligence in this
matter is ruinous to the pronunciation.
THE VOWELS 39
37. Close [a] is represented orthographically by many
signs : a, d, d, e, oi, ot, oie, oy, one, etc. In regard to length,
it follows the seven- vowel rule, i.e. when stressed it is long
before [v, z, 3, j], or final [r]. The circumflexed oi, when
stressed, is also long, if in a closed syllable, as hoUe [bwait].
When unstressed, close [a] is always short, and somewhat
less tense.
38. In the ordinary spelling, close [a] occurs in a variety of
syllables and terminations. These may be reduced to the
following cases :
(i) The letter a :
At the end of a word, as papa [papa], il tomba [toba],
il fera [fara]. Exceptions are fa, la (musical notes),
and beta [beta].
Followed by a mute consonant or consonants, as
chat [Ja], estomac [estoma], drap [dra], je combats
[koba]. There are one or two exceptions, however,
such as climat [klima], chocolat [fokola] (Spanish
word), gars [ga] ; and in the termination -as the
vowel is only close as a rule in bras [bra] and in
verbal endings, as tu auras [ora], tu parlas [parla].
In all closed terminations, except a few special
ones which are referred to under open a (see
next chapter). Thus, nappe, sac, chaque, bague,
chasse, mal, travail, etc.
And generally in all non-final syllables, except in
-ation and a few other cases (see next chapter).
Thus, Canada, animal, madame, papier, casserole, etc.
(2) The letter a, as la, dega ; and the letter a in verbal
endings, as nous donndmes, vous donndtes, il donndi
[donam, donat, dona].
(3) The letter e in a few isolated words, and in the ad-
verbial termination -emment, as femme [fam], solennel
[solanel], nenni [nani], prudemment [prydamd]. In many
of these words, however, such as nenni, hennir, indemnity,
and solennite, the [a] sound is giving way to [e], under
the influence of orthography. Compare (7) below. The
termination -emment, however, remains unchanged, on
account of its constant relationship with -ant, -ent.
(4) The groups oi, 01 (except after r, and in a few isolated
words mentioned in next chapter) . In the seventeenth
40 MANUAL OF FRENCH PRONUNCIATION
century 'they were pronounced [we], and are still so
pronounced in many provinces. In normal French,
however, they now take the sound of [wa], as moi [mwa],
oiseau [wazo], goitre [gwaitr]. We have also close a,
though coming after r, in miroir [mirwair], iiroir [tirwair],
terroir [terwair], Bavarois [bavarwa], and Hongrois
[ogrwa].
(5) The group oy, except after r. As a rule, it only occurs
medially, before an audible vowel, being pronounced
[waj] : loyal [Iwajal], voyant [vwaja], nous noyons [nwajo].
In the few final cases, it is [wa], as in Darboy, Fontenoy,
Troyes.
(6) The group oie when verbal, except after r. It is
sometimes sounded [waj] (not [wa]), as je noie [nwaij],
que je vote [vwaij], je ploierai [plwajre], ploiemeni
[plwajma]. But this is generally regarded as a pro-
vincial ism. •'•
(7) The groups oue, oe. In several words with these
groups, the normal pronunciation used to be [wa], and
this is still retained in some, as couenne [kwan], rouen-
nerie [rwanri], moelle [mwal], moelleux [mwal0], moellon
[mwalo]. But in others, under the influence of ortho-
graphy, the pronunciation [we] is gaining ground from
year to year. Thus, we have jouet [3we], rarely [3wa],
fouet [fwe], rarely [fwa], couette [kwet], rarely [kwat].
In others again, the [we] sound alone is used, as girouette
[3irwet], chouette [fwet], mouette [mwet], rouet [rwe].
In broueite the ou is a full vowel [bruet].^
The same [wa] sound is occasionally heard in souhatt,
souhaiter, but the normal pronunciation is [swe], [swete].
The group oi is pronounced [we], as goHeite [gwelet],
goimon [gwemo], goHand [gwela].
* See footnote to § 41 (7). * Footnote to § 72.
THE VOWELS
EXERCISES
41
a
Sa, ta, va, r6s6da, il porta, il sera
sa, ta, va, rezeda, il porta, il
sara
Plat, j'abats, 6tat, combat, tu vas
pla, 3aba, eta, koba, ty va
Tache, place, canne, salle, nappe,
relaps, jaspe, algue, calme,
halte, ils avalent, valse, drame,
fiacre
taj, plas, kan, sal, nap,
ralaps, 3asp, alg, calm,
alt, ilz aval, vals, dram,
fjakr
Fatal, banal, camarade, cabale
Paris, charity, chapitre, marcher
fatal, banal, kamarad, kabal
pari, Jarite, Japitr, marje
e
oi
oi
oy
oie
Cel^, voil^, c^, hol^, del^
Aimat, aimames, aimates, cher-
chat
Hennir, indemnity, solennit6,
ardemment
Je bois, quoi, soif, avoir, Eloigner
Goitreux, cloitre, cloitrier
Noyer, employer, tournoyer,
soyeux
11 aboie, il nettoie, ils envoient
sala, vwala, sa, ola, dala
sma, emam, emat, Jerja
aniir, edamnite, solanite,
ardama
3e bwa, kwa, swaf, avwa:r,
elwajie
gwatr0, klwa:tr, klwatrie
nwaje, aplwaje, turnwaje,
swaj0
il abwa, il nstwa, ilz avwa
Voilk la femme qui sera k toi.
Va, la dame frappe sur la table.
Le soir nous mangeames avec la pay-
sanne.
11 y a une carafe danoise k ma table.
Madame est malade et marche mal.
vwala la fam ki sara a twa.
va, la dam frap syr la tabl.
l9 swa:r nu masam avek la
peizan.
il i a yn karaf danwa:z a ma
tabl.
madam e malad e marj mal.
CHAPTER VII
OPEN A [a]
39. Open a, as already stated, is essentially the same sound
as in the English words father, rather, psalm, calm, where it
is long. The short form is not found as a rule in Southern
English except in diphthongs, as my [moi], tie [tai], etc., but
it occurs frequently in Scotch and American English, as well
as in French. It is the widest of vowel sounds, requiring the
greatest opening of the mouth ; and also the deepest, requiring
the lowest position of the tongue, which lies flat in the floor
of the mouth and wholly covers it. The difference between
close and open a is seen in such French words as matin and
matin, chasse and chdsse, ma and mat, patte and pate, etc.
The beginner should beware of pronouncing open a like
aw, as in paw, which is really an open o. In some parts of
the English-speaking world we hear such words as father,
can't, pronounced fawther, cawnt. This fault has a much
worse effect in French than in English.
4fO. Open a is represented orthographically by a, a, oi, ot, oe
[wa]. It conforms to the rule of the naturally long vowels ^
[a, o, 0], being short only when final, that is, when in the
last syllable of a word, without a consonantal sound after it,
as degdt [dega], mat [ma]. When non-final, it is usually
long when stressed and half-long when unstressed, as
pate [pait], pate [pa'te].^
41. It occurs generally in the following cases in ordinary
spelling :
(i) The letter a when circumflexed (a), except in verbal
endings. Thus, pale [pail], dpre [aipr], pdtir [patiir].
1 See page 90, § 104 (3).
2 For all practical purposes, half-length may be disregarded and
included under ' short.'
42
THE VOWELS 43
Many of the unstressed cases represent an old -as- now
contracted to a, as tdcher, fdcher, rdler, bdter, hdtir, hldmer,
pdmer, chateau, pdquerette, etc. In bdbord, however, the
a is close, the circumflex being due to a false ety-
mology.
(2) In the termination -as, whether the s be silent or not,
as cas [ka], ias [ta], matelas [matla], atlas [atlais], hilas
[elais], Kansas [kasais]. In hras and verbal endings,
however, the sound is close [a].^
(3) In many cases in the termination -asse, as classe [klais],
Schasse [e/ais], tasse [tais], grasse [grais].^ In Paris at least,
such pronunciation is pretty general, though it appears
uncouth to some French people who do not trouble to
open the mouth so much, and prefer [a]. There are,
however, numerous instances, even in Paris, where -asse
has the vowel close, as crasse (dirt), chasse, masse, cuirasse,
bicasse, crevasse, carcasse, paillasse, brasse, etc. It is
close also in all those words where the suffix -asse takes
a * pejorative ' or unfavourable meaning, as savantasse,
bestiasse, fillasse, etc. Cultivated French speakers, indeed,
are by no means in accord as to the sound of the vowel
a followed by s.
(4) In the group -aille, pronounced [aij], as bataille [bataij],
Versailles [vsrsaij], taille [taij], cisailles [sisaij]. On the
other hand, when this group occurs medially, it is frequently
close, as in ailleurs, bataillon, maillot, assaillir, faillir,
gaillard, caillou, etc. It is also close in midaille, que faille,
que je vaille, qu'il faille. Note that the termination -ail
is always close [aij], as travail [travaij], detail [detaij],
as also are derivatives from it, as travailler [travaje], de-
tainer [detaje]. An exception is rail [raij].
(5) In the group roi, roi, pronounced [rwa], as endroii [adrwa],
octroi [oktrwa], croitre [krwaitr], froisser [frwase].
^ There seems a tendency for it to become close in many other words
in -as, where s is silent, particularly in the endings -las, -nas, -ras, -tas.
For instance, even in Paris, one often hears it close in matelas, chasselas,
cervelas, verglas, ananas, cadenas, embarras, taffetas, galetas, etc.
* Many of such cases are derivatives from words in -as, where the
vowel is open, as grasse from gras, basse from bas, lasse from las, etc.
The open sound is also carried into the verbs, and hence we have it in
such words as amasser, ramasser, passer, trepasser, sasser, ressasser,
tasser, entasser, compasser, damasser, prelasser, etc. See § 43 (i).
44 MANUAL OF FRENCH PRONUNCIATION
In a few isolated words, too, the symbol oi, though not
preceded by r, is given the open sound [wa] by some speakers.
This is so in le hois, hautbois, la boite, foi, mois, notx, pots,
empots, potx, poids, contrepoids, toil, voix. In some of
these cases the open sound serves to distinguish homonyms.
Thus le hois [bwa], but je hois [bwa] ; la boite [bwait], but
il boite [bwat] ; foi [fwa], but fois [fwa] ; mois [mwa],
but moi [mwa] ; toit [twa], but toi [twa] ; voix [vwa], but
je vois [vwa].
(6) In the group roy, pronounced [rwa] when final and [rwaj]
in other cases, as Geoffroy [3Dfrwa], Godefroy [godfrwa],
royal [rwajal], croyons [krwajo].
(7) In the group roie [rwa], and in the termination -oie
in nouns. Thus, proie [prwa], soie [swa], joie [3wa]. When
roie is verbal (the oi representing oy), it is sometimes sounded
[waj], as il foudroie [fudrwaij], ils croient [krwaij], broiement
[brwajma].^
(8) Before a final z sound (i.e. -ase, -az, -aze),^ and in the
terminations -azon, -ason. Thus, case [kaiz], gaz [gaiz],
gaze [gaiz], gazon [gozo], blason [blozo]. An exception is
diapason [djapazo].
(9) In the terminations -ation, -assion, as nation [nasjo],
observation [opservasjS], passion [pasjo]. In many such
cases, however, there is a tendency to close the vowel.
(10) Frequently also in the following terminations, viz.
-abre, -adre, -afre, -avre, -able (not the affix), -rade. Thus,
sabre [saibr], ladre [laidr], Kafre [kaifr], Havre [aivr], sable
[saibl], miracle [miraikl].
(11) In a few cases before final r, as rare [rair], barre [bair],
bar [bair], gare [gair]. In most cases, however, the vowel
is close.
* This pronunciation [waij] or [wa:j] (§ 38 (6)) in the verbal termina-
tions -oie, -oient is regarded as a provincialism. The simple [wa] or
[wa] is the correct pronunciation, but the [j] is usually pronounced in
the case of -ayer and -eyer verbs (§25 (6)), especially in cases where
the y is retained {je paye, je payerai, etc.). In verse the mute e, which
does not count for a syllable in paierai, counts in payerai.
* But not -oise, which is close {bourgeoise, courtoise, danoise, etc.).
THE VOWELS 45
(12) In several isolated words, as flamme, manne, Jeanne,
Anne, Jacques, gars [ga], chocolat, climat, crabe, espace,
esclave, poele [pwail],^ damne [dam], gagne [gaiji], dame,
and derivatives from any of these.
42. While the above rules are generally accepted, it should
be understood that French speech varies considerably on
the subject of the two «'s. It is often a delicate question to
decide whether such and such a word should be pronounced
with [a] or [a] ; and this applies also to the symbol oi. The
sound varies from individual to individual, and depends
largely on the district. It may be said that [a] is more
common in Brittany and in Provence, where such a word as
pate is often pronounced like patte [pat], while [a] is more
general in Normandy and in Eastern France. In any case
where there is doubt or hesitation, the student is recom-
mended to use [a]. 2
1 But poile (frying-pan) is [pwal].
* Many excellent teachers do not recommend the use of [a] in ordinary
speech, and do not regard it as indispensable to French. It is certainly
Parisian, however, and adds beauty to the language.
46
MANUAL OF FRENCH PRONUNCIATION
EXERCISES
a
Acre, cable, patre, dme, bat,
appat, batard, anier, mdter,
platrer, patee
a:kr, ka:bl, pa:tr, aim, ba,
apa, bata:r, anje, mate,
platre, pate
-as
Bas, pas, amas, appas, repas,
gras, vasistas, Leonidas,
Stanislas, as
ba, pa, ama, apa, r9pa, gra,
vasistais, leonidais, stanis-
la:s, a:s
-asse
Lasse, passe, amasse, casse,^
nasse, basse, ressasse, pr6-
lasse, Parnasse
lais, pais, ama:s, kais, najs,
ba:s, rasais, prelais, par-
na:s
-aille
Paille, braille, rocaille, fian-
9ailles, mangeaille, volaille,
marmaille, caille
pa:j, braij, rokaij, fjasaij,
ma'^aij, volaij, marmaij,
ka:j
-roi
-roi
-roy
Froid, croit, croix, droit, pa-
roisse
Croit, croitra, surcroit, je crois
Leroy, croyez, broj^er, proyer,
broyeur
frwa, krwa, krwa, drwa,
parwas
krwa, krwatra, syrkrwa, 39
krwa
larwa, krwaje, brwaje,
prwajc, brwajoeir
-roie
-oie
Broie, Troie, que je croie, fou-
droiement
La voie, foie, oie
brwa, trwa, ka 39 krwa,
fudrwama
la vwa, fwa, wa
[z]
Vase, phrase, rase, base, ecrase,
stase, phase, topaze, rason
va:z, fraiz, ra:z, ba:z, ekraiz,
sta:z, fa:z, topa:z, razo
-ation
Exploration, gradation, ration,
occupation, spiration, stagna-
tion
eksplorasJD, gradasjS, rasjo,
okypasjo, spirasjo, stag-
nas j 5
Macabre, calabre, glabre, cadre,
madre, balafre, navre, ca-
davre, fable, jable, diable,
accable, oracle, rafle
makaibr, kalaibr, gla:br,
kaidr, maidr, balaifr,
na:vr, kndaivr, fa:bl,
3a:bl, djaibl, akaibl,
oraikl, ra:fl
-r
Mare, contrecarre, mars, jarre
mair, kotrgkair, mars, 3a:r
Je crois que Jeanne et Jacques sont las.
Le tailleur a gagne la medaille.
Geoffroy a pris trois tasses de the.
L'ane a tache d'ecraser les vases.
Jadis on faisait de la sole dans la Chine.
39 krwa k9 3am e 3a:k s5 la.
l9 tajoeir a gajie la medaij.
3ofrwa a pri trwa ta:s d9 te.
la:n a taje dekraze le va:z.
3adis 5 f9zs d9 la swa da la
Jin.
' Open when denoting 'breakage,' or a printer's 'case,' but some-
times close when meaning ' cassia/ or ' basin.'
CHAPTER VIII
UNSTRESSED A
43. When unstressed, the vowel a, whether close or open,
is generally pronounced with a little less tension of the
muscles. The following points regarding it should also be
noted :
(i) Derivatives generally follow the root- word or stem,
unless some modifying influence arises to change the sound.
Thus, chasser [Jase] from chasse [fas], but passer [pase] from
passe [pais] ; travailler [travajs] from travail [travaij], but
hrailler [braje] from braille [braij] ; soi-disant [swa-diza] from
soi [swa], but soierie [swari] from soie [swa]. We give here
some examples in connexion with open a (see Footnote 2,
page 43) :
stem
Derivative
Stem
Derivative
Stem
Derivative
[open)
[open)
{open)
[open)
{open)
{open)
Sable
sablon
Accable
accabler
Phrase
phraser
Gaze
gazeux
Navre
navrer
Ecra^e
ec raser
Diable
endiabler
Sasse
sasser
Jase
jaser
Flamme
enfiammer
Lasse
lasser
Noix
noisette
Gare
garer
Rase
raser
Gazon
gazonner
xMadre
madre
Extase
extaser
Blason
blasonner
Delabre
delabrer
Vase
vaser
Espace
espacer
Cadre
encadrer
Case
caser
Clame
clamer
(2) In some cases, however, difference of stress acts as a
modifying influence, and changes the quality of the vowel
from [a] to [a]. The further the vowel is from the tonic
stress, the greater is the tendency to this. Thus we have
[a] in hajricade, grasseyer, fahuliste, damnation, cadsivereux,
iriflammation, diabolique, cadran, etc., in spite of [a] in
the root-words. The same tendency appears in compound
words; thus the open [a] in passe becomes quite close in
47
48
MANUAL OF FRENCH PRONUNCIATION
passementerie, and even with some speakers in passeport and
passepoil. Other examples are :
a
a
a
a
Acre
acrimonie
Infame
infamie
Fracas
fracasser
Jeanne
Jeanne tte
Jacobin
Embarras
embarrasser
Jacob
Classe
classique
Poix
poisser
Case
casamer
Bois
Dois6
Affres
affreux
Grace
gracieux
Anis
anisette
Cas
cas de conscience
Jacques
Ja^quard
Bas
bas de soie.
(3) There are many words more or less isolated or un-
derived which have [a] in the unstressed syllable. Among
these may be mentioned the following : Calais, magon, limagon,
colimagon, magot, maffle, maquis, jadis, brasier, cassis, pasieur,
praline, scabreux, anis. There are also many cases where the
vowel is open before a z sound, as basalte, basane, jaseran,
masure, mazette, Jason, etc. Compare § 41 (8).
(4) In some unstressed syllables [a] or [a] is replaced by a
sound intermediate between the two. In other words, just
as there is a middle e, so there is a middle a. It is less tense
than the close sound, the mouth is a little more open, and the
tongue a trifle lower, but not so much as to sound [a]. It
generally occurs before r, in a pretonic syllable, i.e. in
an unstressed syllable immediately preceding the stressed
one. It is difficult sometimes to know whether to class it
with [a] or with [a], as it seems to vary between the two.
Some phoneticians write it as [a]. It approaches [a] in such
words as carreau, carrosse, carre, carotte, baron, garenne, marron,
parrain, marraine, sarrau, etc., where the r is mostly doubled.
But it comes nearer to [a] in most words, especially where
another consonant follows [r], as mardi, partir, tarder, marcher ^
marmite, charpie, darder, artilleur, etc. With a few exceptions,
therefore, it may be included under the phonetic symbol [a],^
to save multiplicity of signs.
^ In English, on the other hand, [a] predominates before [r].
CHAPTER IX
CLOSE 0 [o]
44. There are two o's, the one close and the other open.
They are distinct in sound from each other, though repre-
sented generally by the same orthographic symbols {o, 6, au,
oa, etc.). The correct enunciation of the two sounds should
occasion no difficulty to the student. The close sound [o]
is the Scotch or Northern English pronunciation of o in
such words as note, wrote, coat, where it is long, and notation,
rotation, coterminous, where it is short. The open sound [o],
on the other hand, is found long in shawl, wrought, caught,
and short in block, rot, cot.
45. Referring to the triangle of vowels in § lo, it will be
noticed that on passing from [a] to [o] the tongue bunches
up a little towards the back, the mouth closes a trifle, and
the lips become somewhat rounded and projected. These
characteristics are all intensified in the case of close [o], for
which— in French at least — the rounding and pushing out
of the lips is an important matter that must not be neglected.
The muscles, too, are held tenser for [o] than for [o],
46. The close sound, as stated, is found purest in Scotland,
where it is a monophthong. In the South of England it is
replaced by a diphthong ([ou], etc.), the second element of
which is generally oo (phonetically [u]). This is due to want of
lip-rounding and lack of tension. The vocal organs are not
held sufficiently fixed in their position, with the result that
the vowel, especially if it be long, does not maintain its quality
but tends to glide into another. The simple remedy is to
round and project the lips sufficiently, and keep the
muscles tense, thus preventing any gliding movement so
long as the sound continues. Such a word as cote in French
must not be pronounced [kout] or [kout], but [koit].
47. In regard to length, close o follows the rule of the
n 49
50 MANUAL OF FRENCH PRONUNCIATION
naturally long vowels [a, o, 0], already referred to under [a]
(§ 40). It is generally found as follows in ordinary spelling :
(i) In all cases of final o, that is, when no consonantal sound
follows, as dos [do], galop [galo], numiro [nymero]. The
vowel of trop, however, becomes open in liaison, as trop
aimable [tropsmabl], and some people open it always.
(2) All cases of circumflexed o, as iot [to], drole [droil], fantome
[fatoim], noire [noitr], Saone [soin], oter [ote], cloture [klotyir].
But it is frequently open in aumone.
(3) In the termination -otion, as Smoiion [emosjo], notion
[nosjo], potion (posjo].
(4) The vowel o before [z], 2iSrose [roiz], poser [poze], explosif
[eksplozif], groseille [grozsij], Buloz [byloiz]. There are a
few exceptions : philosophe [filozof], myosotis [mjozotiis],
losange [bzai3], cosaque [kozak], hosanna [ozanna], and
moza'/que [mozaik], together with words in -osition (especially
proposition) and those which commence with pros- {prosdique,
prosilite, etc.).
(5) In some cases in the terminations -ome, -one. In olden
times these terminations were always close. They re-
presented learned words, as distinct from the terminations
-omme, -onne, which occurred in words in common use.
In most cases they are now pronounced open [om], [on].
But in a few isolated words one hears [oim], [om], as arome,
atome, axiome, brome, chrome, idiome, tome, zone, lazarone,
cyclone, icon. In others again there is hesitation between
[0] and [o], as in gnome, hippodrome, amazone, atone,
polychrome.
(6) In some cases in the termination -osse. In most cases
this is pronounced [os], but [ois] is common in dosse [dois],
fosse [fois], grosse [grois]. Grosse (twelve dozen) is generally
[gros].
(7) The groups eau, au, except before r and in a few
isolated words. Thus, maux [mo], anneau [ano], gaule
[goil], cause [koiz], heaume [oim], auteur [otoeir], nautik
[notik]. Lausanne J however, is frequently [lozan].
In the many cases of unstressed au the words include
those in which this symbol represents an old French
diphthong now contracted, as aucun, aussi, autant, chauffer,
fausser, hauteur, sauter, etc. They also include a large
number of learned and recent words, as auspices, austere,
THE VOWELS
51
austral, authentique, auiographe, automobile, autriche, caution,
etc.i
(8) In a few isolated words, as obus [obyis], odieux [odj0],
odeur [odoeir], momie [momi] and its derivatives.
' Some of these words, however, are frequently pronounced with
[o] when the vowel is followed by two consonants, as [ospis, ostsir, astral],
etc., and even in several other cases {autoriser, autoviU, authentique^
auguste, etc.).
EXERCISES
Final
Alto, folio, mot, clos, gros, heros,
pot, dos, idiot, devot, trot,
alto, foljo, mo, klo, gro, ero,
po, do, idjo, devo, tro,
3igo
6
Rot, depot, aussitot, il clot,
prevot, cote, hote, rode,
diplome, pole, Rhone, con-
troler. geolier, cloturer, tot-
fait
ro, depo, osito, il klo, prevo,
ko:t, o:t, roid, diploim,
po:l, rom, kotrole, 3olje,
klotyre, tofe
-otion
Commotion, lotion, devotion,
motion
komosjo, losjo, devosjo,
mosjo
z
sound
Chose, dose, prose, pose, close,
Gaidoz, Joseph, oser, position,
positif, roseau
Jo:z, do:z, pro:z, poiz, klo:z,
g£do:z, 30Z8f, oze, pozisjo,
pozitif, rozo
au
Chapeau, etau, defaut, il vaut,
il faut, faute, paume, epaule,
aube, auge, badauderie, sau-
mon, saulaie, vautour
Japo, eto, defo, il vo, il fo,
fo:t, po:m, epo:l, o:b,
0:3, badodri, somo, sols,
votu:r
Le chateau est pos6 au haut des Vosges.
Claude a saute dans I'eau apres
I'agneau.
Le pauvre idiot est sous le gros saule.
L'autre veau a saute par-dessus nos
pots.
Les rosiers du chateau sont beaux.
la Jato 8 poze 0 0 de vo:3.
klo:d a sote da lo apre 1 apo.
la povr idjo 8 su la gro soil,
lotr vo a sote par-dasy no
po.
le rozje dy Jato s3 bo.
CHAPTER X
OPEN 0 [o]
48. Open o, as already mentioned, is practically the same
sound as in the words shawl, tortoise (long), and block, waddle
(short). Attention should be paid to the clear, precise
enunciation of the vowel. Some affected people speak of
* culleges,* ' schullars,' ' cunscience.' Such deviations from
the true sound are quite unintelligible when embodied in
French words.
49. As to length, open a follows the seven-vowel rule, i.e.
when stressed, it is long before [v, z, 3], or final [r] (it
does not occur before [j]^), as ove [oiv], loge [1013], mori [moir],
Badajoz, [bada3oiz].2 When unstressed, it is almost in-
variably short, as joli Ijoli], solide [soUd], local [lokal], logeiir
[b3oeir].
50. The student should have no difficulty in knowing when
this vowel occurs in ordinary spelling. If he remembers the
particular cases of close 0 referred to in the previous chapter,
he will understand that in all other cases the symbols 0, 6,
au, etc., represent open 0. The following cases of it, however,
may be specially mentioned ^ :
(i) The letter o (not 6) in closed syllables (except before
[z], and in some cases of -ome, -one, -osse). Thus, posie
[post], alors [abir], golfe [golf], nostalgie [nostal3i].
In the termination -os, where the 5 is sounded (mostly
foreign words and names), the vowel is generally pronounced
^ Except in o'il [o:j], old French for oui.
2 The sound o before [z] is close {§ 47 (4)), with the exception of one
or two proper names, as Badajoz, Booz [boo:z], Berlioz [berljoiz].
2 In addition to these cases, the close 0 opens somewhat in liaison (as
in pot-ct-l'eau, pot-au-lait, mot-il-mot, croc-en-jambe, trop aimable, etc.),
coming really under (2).
52
THE VOWELS 53
long, as pathos [patois], rhinoceros [rinoserois], Argos [argois].
In Calvados the 0 is close.^
(2) The letter o (not 6) in open, non-final syllables, as
modeste [modest], voler [vole], ovale [oval], igotste [egoist],
poeme [poeim], zoologie [zoolo3i], morille [moriij]. But it is
close in boa [boa], and oasis [oaziis] and a few more.
In the prefix co-, the vowel is pronounced close by some
people, especially before [a] or [e], as coaguler [koagyle],
coalition [koalisjo], coexister [koegziste].
(3) The group au before [r], as taure [toir], Maure [moir],
laurier [lorje], faurai [3ore], Aurillac [orijak].
Also in a few other words, as Paul, Paule (generally),
Auch [oj], augmenter (and derivatives), Auxerre [oseir],
and sometimes in sauf (preposition) and mauvais. In the
case of aumone, when the 6 is pronounced open, the au is
generally open also, owing to vocalic assimilation [omon] ;
but in aumonier, though the 0 may still be open, the au is
pronounced close [omonje].
(4) In the termination -um (mostly Latin words), as decorum
[dekorom], album [albom], opium [opjom], as well as
medially in some compound words (triumviral, circum-
navigation, etc.).
(5) The group 00 in looch [lok], and alcool [alkol] and derivatives.
^ Many French speakers pronounce the 0 close in numerous other
cases of -os ; but except where the words have an w in Greek, it is better
to adopt the open sound.
54 MANUAL OF FRENCH PRONUNCIATION
EXERCISES
o
o
Robe, brioche, roche, ode, etoffe,
dogme, propre, 6poque,
octobre
Soleil, violet, volatile, comique,
cochon, oratcur, heroique,
local
rob, brioj, roj, od, etof,
dogm, propr, epok,
oktobr
soleij, vjole, volatil, komik,
kojo, oratoe:r, eroik, lokal
co-
Coasser, coactif, coeternel,
coequation
koase, coaktif, koeternel,
koekwasjS
au
Centaure, saiirc, Faure, Laure,
epidaure, aurore, je saurai,
aureole, auricule
sato:r, so:r, fo:r, loir, epidoir,
oro:r, 39 sore, orcol,
orikyl
um
Rhuni, pensum, laudanum,
museum
rom, pesom, lodanom,
myzeom
Paul apporte des coqs et des cigognes.
11 donne quatorze pommes k notre
bonne.
Cet homme adopte des dogmes ortho-
doxcs.
En aiitomne I'orge est color^e comme
Tor.
L'album est dans notre coflfre de bord.
pol aport de kok e de sigop.
11 don katorz pom a notro
bon.
set om adopt de dogmoz
ortodoks.
an oton lor3 e kolore kom
lo:r.
lalbom 8 da notro kofra do
boir.
CHAPTER XI
UNSTRESSED 0
51. When unstressed, the vowel o, like the other vowels, is
pronounced with a little less tension of the muscles. The
following points should also be noted :
(i) Derivatives generally follow the root- word or stem,
unless they become altered through some outside influence.
Thus, dossier [dosje], adosser , endosser homdosse [dos], grossir
[grosiir], grosseur from gwsse [grois], fosse [fose] from fosse
[fois],^ zone [zone] from zone [zom], enjoler from geole, enroler
from role, Pauline [polin] from Paul [pol] .
(2) Vocalic assimilation (§ 29) occurs in some words, especi-
ally where the same syllable is repeated, as bobo (bobo],
coco [koko], rococo [rokoko], dodo [dodo], gogo [gogo], lolo
[lolo].
(3) There is a middle o, intermediate between the close
and the open sound. Phonetically, it is generally represented
by [0]. The back of the tongue is a little lower than for close
[o], but not so low as for [o] ; and the mouth is a little more
open and the muscles slightly relaxed. It frequently takes
the place of [o] or [o] in unstressed syllables, and with only
the two symbols it is sometimes difficult to know with which
to class it. Probably it is better to class it in most cases
with [o]. It occurs in some derivatives from 6. Thus we
have close [o] in the root- words cote, rot, pole, diplome, but
middle [0] in coteau, cotelette, roti, hotelier, hopital, polaire,
diplomate. It also occurs frequently in ordinary open syl-
lables, as poteau, prononcer, comment, noter, rosace, roseole,
etc.
52. The student should not trouble himself much over
middle [e, a, 0]. In nearly all cases they may be neglected,
^ The 0, however, is sometimes open in fossette, generally in fossoyer,
fossoyeur, and always in fossile.
55
56 MANUAL OF FRENCH PRONUNCIATION
and their place taken by the close or open sound, whichever
is the nearer, without injury to the pronunciation. If the
beginner will only set himself to surmount the difficulties in
the way of a correct and clear enunciation of the eight standard
vowels, he may ignore the discrimination required for the
middle sounds until later.
CHAPTER XII
CLOSE [u]
53. The open [u] is frequently heard in EngHsh (as in full),
but is not found as a rule in French except in the middle
provinces. The close sound of the vowel prevails, being
the same as that in the English words rude, rule (long), or
good, July (short). Close [u] is never represented, however,
in ordinary French spelling by u, but solely by ou, cU, as jour,
rouge, nous, couter, etc.
54f. For the proper enunciation of close [u], the tongue
rises towards the back still more than for [o], the tip going
back almost to the middle of the mouth-floor (see diagram, § 8),
There is also a greater projection and rounding of the lips,
the orifice made by them being only large enough to allow
the insertion of an ordinary lead pencil. The muscles of the
tongue and lips are held very tense.
55. Many English-speaking students pronounce the French
[u] badly, because they do not push out the lips enough (this
projection being uncommon in English), and do not hold the
muscles sufficiently tense but allow them to relax before the
sound finishes. The French [u], it should be remembered,
is a vigorous firm sound. It should not be allowed to dis-
appear, as it tends to do in rapid, careless speech. Thus, we
hear vous avez (vzave], voulez-vous vous taire [vlevuvtEir],
tout-d-fait [t ta fs], tout d Vheure [t ta loeir], etc. This is a
negligence that should not be imitated.
56. In regard to length, [u] follows the seven-vowel rule,
i.e. when stressed, it is long before [v, z, 3, j], or final [r], as
irouve [truiv], blouse [bluiz], rouge [ruij], mouille [muij], court
[kuir]. It is also long in tous (the pronoun) when under stress
(as in pour tous), and generally in coule, route, ecroule (see
§ 104 (4) (c)). In all other cases it is short, even though
57
58
MANUAL OF FRENCH PRONUNCIATION
circumflexed, as in mou, joug, loup, cout, gout, mout, saoul [su].
When unstressed, it is short and less tense, becoming often
half-open. Thus, houton [buto], coucou [kuku], couleur [kuloeir],
ouragan [uraga], ouhli [ubli], douter [dute], troupeau [trupo],
moucher [mu/e], gouter [gute].
57. The word aout is correctly pronounced [u], but the
archaic pronunciation [au] is again becoming frequent in
conversation, and even [ut] and [aut] are sometimes heard.
EXERCISES
Long
Louvre, louve, ouvre, couve,
couse, spouse, douze, ven-
touse, bouge, rouille, souille,
grenouille, four, amour, lourd,
secours
luivr, lu:v, u:vr, ku:v, ku:z,
epuiz, du:z, vatuiz, bu:3,
ru!J, su:j, granu:j, fu:r,
amu:r, lu:r, sakuir
Short
Vous, joue, choux, soul, caillou
goutte, bouc, poule, mouche,
lourde, courbe, courte,
fourche, journee, boucher,
bonder, gouter
vu, 3U, Ju, su, kaju, gut, buk,
pul, muj, lurd, kurb, kurt,
fur J, 3urne, buje, bukle,
gute
11 laboure pour nous tous les jours.
Lcs poules courent sur la route.
Le bouc broute sous la voute.
Voulez-vous souper sur la mousse ?
La blouse de son 6pouse est rouge.
il labuir pur nu tu le 3u:r.
le pul ku:r syr la rut.
l9 buk brut su la vut.
vule vu supe syr la mus ?
la bluiz d9 son epuiz e ru:3.
CHAPTER XIII
THE MIXED VOWELS
58. We have now described the eight fundamental vowel
sounds, and proceed to the three mixed ones [y, 0, oe].
They are so called because, although they make a simple
impression on the ear, they are formed by the union of two
of the fundamental ^sounds. The lips take the opening re-
quired for the one, while the interior of the mouth assumes
the shape belonging to the other, the result being a mixed
sound not found in English. Thus, referring to the triangle
in § 10, it is evident that we may put the lips in the rounded
position required for [u] (as in English rude), as if we were
actually going to sound this vowel, and while keeping them
rounded sound the vowel [i] (as in English machine), i.e. move
the tongue forward to the position required for the corre-
sponding front vowel. We thus get the mixed sound [y],
as in the French sur, tune, tu, etc. Similarly, we have a
sound [0], made up of the lip position of [o] and the tongue
position of [e], as in deux, feu, etc. The third mixed vowel
[oe] is formed from the lip position of [o] and the tongue posi-
tion of [e], as in leur, neuf, sceur, etc. The triangle may thus
be more fully arranged as follows, the brackets round a vowe!
denoting rounding and pushing out of the lips ;
(CZo5e;>s- (y)- ^~ --/^^M
(MiddmS^
(Oper^cu
60
6o MANUAL OF FRENCH PRONUNCIATION
The Vowel [y]
59. This vowel, as above stated, is produced simply by
rounding the lips as for [u], i.e. rounding them tightly as
for whistling, and then sounding [i]. The vowel sound in
the Scotch guid is somewhat similar, but is open, whereas the
French sound is close and tense. The sound is so frequent
that complete mastery over it is required, if one is to produce
it easily and quickly. It should be practised thoroughly,
until the muscles of the mouth are accustomed to the com-
bination. Constant and careful repetition of the three sounds
[i, u, y], backwards and forwards, will considerably help.
Care should be taken (i) that the lips are sufficiently projected
and rounded, and (2) that the movements of the lips and those
of the tongue are made simultaneously and not in succession.
60. Some difficulty may be experienced at first in pro-
ducing the sound clearly after [r], [s], or [t] (as in rue, su,
tu), but persistent practice will succeed. In the case of [t],
the difficulty with beginners is largely due to their forming
the consonant in the English instead of the French way.
For the English t the tongue tip strikes the hard palate a
little behind the upper teeth, but for the French t it must strike
it immediately behind (§ 150). Practice should be made with
such words as Huve, astuce, turc, tube, tuf, etude, laitue, tu,^ etc.
61. When stressed, the vowel [y], like the two from which
it is formed, is long before [v, z, 3], or final [r], and generally
also in the termination -us (mostly in classical words) where
[s] is sounded. Thus, cuve [kyiv], ruse [ryiz], juge [5yi3],
voiture [vwatyir], Venus [venyis], omnibus [omnibyis]. In
all other cases, whether stressed or unstressed, it is short, even
though circumflexed, as vu [vy], juste [3yst], flUte [fiyt],
bUche [byj]. When unstressed, it is uttered with less tension,
becoming frequently half-open, as buvard [byvair], musique
[myzik], murmur er [myrmyre].
62. The sound is generally represented in ordinary spelling
by u. There are some cases denoted by u, as Esail [ezay],
Danaus [danayis], and a few by ue (after g only), as cigue
^ In common speech, the pronoun tu sometimes loses its vowel, as
Tas pas compris ? Que Ves bite t
THE VOWELS
6i
[sigy], aigue [egyj.^ The symbol eu also takes the sound [y]
in gageure [gajyir],^ vergeure [vEr5yir], mangeure [ma3yir],
and in the past participle, past definite, and imperfect sub-
junctive of avoir, as eu [y], feus [3y], tu eus [ty y], il eut [il y],
nous eumes [nuz ym], vous eutes [vuz yt], Us eurent [ilz yir], etc.
In chut the u is silent (/it), unless necessary for rhyme,
or unless to designate the interjection, as : On entendit plusieurs
chut [fyt].
^ The diaeresis over the e is merely to signify that the u is sounded.
' Under the influence of orthography, the pronunciation [ga3oe!r] is
becoming frequent.
EXERCISES
Long
Vesuve, refuse, deluge, azur,
etuve, accuse, refuge, mur
vezyiv, rafyiz, dely:3, azy:r,
etyiv, akyiz, r9fy:3, my:r
Short
ru
su
tu
Chute, culte, nul, luxe, usurpe,
brume, urne, absurde, cruche,
briile, affut, cru, tu, du
fyt, kylt, nyl, lyks, yzyrp,
brym, yrn, absyrd, kr>'J,
bryl, affy, kry, ty, dy
Russe, rude, truffe, ruche
Sue, massue, re9u, supposer
Tulipe, sculpture, tulle, tumulte
rys, ryd, tryf, ryj
syk, masy, rasy, sypoze
tylip, skylty:r, tyl, tymylt
-us
Cresus, Bacchus, Festus,
Taurus, crocus, hiatus, lotus,
prospectus
krezy:s, baky.s, festyis,
toryis, krokyjs, jatyis,
lotyis, prospEktyis
eu
Que j'eusse, qu'il eut, qu'ils
eussent
k9 3ys, kil y, kilz ys
Jules eut une culbute sur la tribune.
J'ai vu plus d'une buche dans la cuve.
Le cur6 aper9ut la perruque du Juge.
Le due ne fume plus dans sa voiture.
11 eut les muscles durs et tordus.
3yl y yn kylbyt syr la tribyn.
3e vy ply dyn byj da la ky:v.
l9 kyre apsrsy la peryk dy
la dyk na fym ply da sa
vwatyir.
il y le myskla dy:r e tordy.
CHAPTER XIV
THE VOWEL [0] (CLOSE eu)
63. This sound is even more difficult to a beginner than [y].
It is sometimes stated in grammars to be like u in the English
word fur, but this is a false analogy, only leading to confusion.
There is no corresponding sound in either English or Scotch.
The sound is formed by combining the rounded lip position
of close [o] with the tongue position of close [e], i.e.
pronounce [o] (as in English tone), taking care to project and
round the lips, and then, without altering the position of
them, sound [e] (as in English case). The result will be a
peculiar resonance coming from just inside the lips, and
forming a perfect French [0]. The usual fault in the case of
English-speaking students is that they fail to advance and
round the lips sufficiently, and they thus sound a kind of
close [e]. The only remedy is practice. Exercises with
[e, o, 0] in succession, backwards and forwards, should be
diligently employed until the sound is under perfect control.
64. This sound is short only when final in pronunciation,
thus following the rule of the naturally long vowels [a, o, 0],
referred to under [a] (§ 40). It is represented in ordinary
spelling by eu, eu, eue, ceu. It occurs as follows, and may
thus be distinguished from the open sound (as in heure, sceur,
etc.) described in next chapter :
(i) When eu is final in pronunciation, as jeu [30], deux
[d0], hleus [bl0], nceud [n0], monsieur [m9sj0] ; and in deriva-
tives from such words, as deuxieme [d0zjsm], bleudtre [bl0aitr],
lieutenant [1 j 0tna] .
The only exceptions are certain forms of avoir, which
are pronounced \y] (§ 62).
(2) In the terminations -euse, -eute, -eutre, and deriva-
tives, as creuse [kr0iz], creuser [kr0ze], Meuse [m0iz],
62
THE VOWELS
63
Smeute [em0it], imeuter [em0te], neuire [n0itr], neutraliti
[n0tralite]. Also in Greek names in -eus, as Zeus [z0is],
Orpheus [orf0is].
(3) In a few isolated words (including their derivatives) ,
such as jeune, jeudi, meule, meunier, pentateuque (learned),
pseudonyme, neume}
Dijeuner, however, is frequently pronounced open
[de3oene], and even [de30ne] and [de3ne] are popular. In
the case of meule, most teachers recommend [0] when the
word means ' haystack,' but [oe] where it means ' millstone,'
as a distinction of some kind exists in most French dialects.
(4) The prefix eu-, as eucalyptus [0kaliptyis], Eulalie [0lali].
Exceptions are Europe, and sometimes Eugene, Euginie,
in which it is open, though it tends to be close in the last.
^ There are also a few other words of a learned nature, such as
feudiste, deuteronome, ichneumon, teuton, and derivatives, together with
some proper names {Beuchot, Ceuta, Deucalion, Neuilly, etc.), and words
in -eiitique, -eumatique.
EXERCISES
Final
Feu, feutier, lieu, lieutenant,
queue, queuter, pneu, pneu-
matique, gueux, gueuserie,
Dieu, oeufs, boeufs
f0, f0tje, lj0, lj0tna, k0,
k0te, pn0, pn0matik, g0,
g0zri, dj0, 0, b0
-euse
rente
-eutre
Creuse, creuser, chartreuse,
berceuse
Meute, ameuter, ameutement,
emeutier
Feutre, calfeutrer, feutrage,
pleutre
kr0:z, kr0ze, Jartr0:z,
b8rs0:z
m0:t, am0te, am0tma,
em0tje
f0:tr, kalf0tre, f0tra3,
pl0:tr
Isolated
Ameulonner, emmeuler, meu-
nerie
am0lone, am0le, m0nri
eu-
Euphonie, eulogie, Euclide,
Euphrate
0foni, 0lo3i, 0klid, 0frat
Adieu, monsieur ! Dieu le veut.
Ces yeux-ci sont plus bleus que ceux-la.
Je veux deux oeufs pour ces gueux.
Ceux qui s'ameutent sont facheux.
Le meunier veut jeuner jeudi.
adj0, m9sj0 ! dj0 la V0.
sez J0si so ply bl0 ka sola.
39 V0 d0z 0 pur se g0.
S0 ki samoit so fajo.
la m0nje V0 30ne 30di.
CHAPTER XV
THE VOWEL [oe] (open eu)
65. To produce this sound one must advance and round
the lips as for open o [o], and while keeping them in this position
sound [e], i.e. dispose the tongue for the corresponding front
vowel. The result is the mixed vowel [oe], as found in such
words as heure, peur, sceur, etc. Care must be taken not
to round the lips as much as for close [o]. A very slight
rounding will do, with the mouth fairly wide open ; and the
muscles should not be held so tense, as both [o] and [e] are
somewhat relaxed sounds and require no great effort either
alone or mixed. By attention to this, there will be no
confusion between this sound and the previous one [0],
which has the lips more rounded and the muscles tense.
Beginners who are unable to make any distinction should
practise words containing both sounds, such as masculines
in -eur (which have [oe]), and their feminines in -euse (which
have [0]). Thus, acheieur, acheteuse ; danseur, danseuse ;
menteur, menteuse ; causeur, causeuse.
66. This sound is represented in ordinary spelling by eu,
ceu, ue, ce. For length, it comes under the seven-vowel rule,
i.e. when stressed, it is long before [v, j], or final [r], and
short in other cases, a.sfleuve [ficeiv], feuille [foeij], peur [pceir],
peuple [poepl]. When unstressed, it is always short and
uttered with somewhat less tension, as jeunesse [3oen8s],
fleurdelise [floerdalize].
67. It is never found in open final syllables (these being
sounded [0]), but only in closed ones ; but in non-final
syllables, there are several open cases. There need be no
difficulty, however, in knowing when the sound occurs.
If the four classes of [0], mentioned in the previous chapter,
be remembered, it will be understood that in all other cases
THE VOWELS
65
the symbols eu, ceu represent [oe], and that ue} ce always do
so. Thus we have [oe] in such words as the following :
peuvent, jeune, seul, veulent, neuf, meuble, veuvage, ecueil,
bouvreuil, feuille, treuil, etc.
68. It should be noted that, while we have [oe] in ceuf and
h(zuf [oef, boef], we have [0] in the plurals ceufs [0], hceufs [b0],
as the / is sounded only in the singular. Vocalic assimilation
(§ 29) also tends in two or three cases to change [oe] into [0].
'Thus we have heitgle [boegl], heuglement [bceglama], but
beugler [b0gle] ; abreuve [abroeiv], abreuvoir [abroevwair],
but abreuver [abr0ve]. On the other hand, peu opens a little
in a peu pres.
^ Ue occurs in words in -cueil, -gueil, where this older form of writing
[oe] has been retained to preserve the hard sound of c and g.
EXERCISES
cu
Neuve, veuve, epreuve, cou-
leuvre, deuil, veuille, fauteuil,
seuil, fleur, ardeur, hauteur,
ailleurs
noeiv, voeiv, eproeiv, kuloeivr,
doe:j, voe:j, fotoe:j, soe:j,^
floeir, ardoeir, otoe:r, ajoe:r»
Heurter, meurtrier, pleurant,
Europe, heureux, peuplade,
veuillez, Fleury
oerte, moertrie, ploera, oerop,
oer0, poeplad, voeje, floeri
oeu
Cceur, soeur, oeuvre, oeuvrer
koe:r, sce:r, oeivr, oevre
ue
Orgueil, orgueilleux, cueille,
cueillir, cercueil, accueil,
accueillir, longueil
orgoeij, orgoej0, koe:j, koejiir,
serkoeij, akoeij, akoejiir,
15goe:j
ce
ceil, oeillade, oeillet, oeillere
oe:j, oejad, oejs, oeje:r
Leurs jeunes soeurs m'ont accueilli.
Les veuves veulent demeurer jeunes,
L'ardeur de leurs coeurs les rend
heureux.
Leur honneur seul est leur oeuvre.
Les aveugles ne peuvent cucillir de
fleurs.
loer joen soe:r mot akoeji.
le voe:v vcel damoere 3oen.
lardoeir da loer koe:r le ra
oer0.
loer onoeir soel e loer oe:vr.
lez avoegla na poe:v koeji:r da
floeir.
CHAPTER XVI
E MUTE or 'E CADUC
69. This is the indeterminate vowel occurring in such
words as le, me, te, se, que, etc., and represented phonetically
by [9]. It is somewhat similar in sound to the e in the English
words over, taken, and to the letter a in about, again, sofa.
For the French sound, however, the lips must be a little
rounded and projected. It is thus a mixed sound, like
[y, 0, oe], and its correct place is on the triangle somewhere
near [oe]. Some phoneticians place it between [0] and [oe],
thus regarding it as a mixture of middle [e] and middle [0],
while others place it a little below [oe], regarding it as [oe]
relaxed [de]. We have placed it in the latter position (see
§ 58), but would draw attention to the main essential for its
correct enunciation, viz. a slight lip-rounding. The sound
is so common in French that it must be thoroughly mastered
from the outset. The sound of e in over should be taken
and practised with lips as described.
This indeterminate sound is simply the decayed form of
earlier distinct vowel-sounds, which became gradually
neglected and were * skipped over ' because they were un-
stressed. In numerous cases, it remains in the language as a
mere graphic survival, no longer pronounced : (i) In the
verbal terminations -aient, -oient, -tent, etc. (2) In those of
futures and conditionals {-erai, -erais) after a vowel, as jouerai,
flierais. (5) -In nouns in -erie after a vowel, as flouerie,
feerie. (4) In the ending -gue, where the diaeresis merely
indicates that the u is sounded, as aigue, cigue. (5) In
such words as Caen, Madame de Stael, Saint-Saens
[ka, stdl, SEsais]. (6) In all cases when it is final and
preceded by a vowel or [j] (-ate, -aye, -ee, -ie, etc). In
such cases as these last, it is pronounced in singing (see
§ 105 (2)).
66
THE VOWELS 67
70. Apart from such cases as those just mentioned, it is
generally sounded (always short), though it has a habit of
becoming almost silent or disappearing altogether, which has
led French phoneticians to call it ' e caduc ' (' deciduous e ').
The cases of this silence or disappearance (Elision) are con-
sidered in Chapter XXXI. When sounded, it should be
distinctly heard, and when stressed in any way (under
tonic accent or emphasis) it becomes [oe], as prends-le [praloe],
fais-le entrer [fdoe atre], oui, je viendrai [wi, 306 vJEdre].^
In ordinary spelling it occurs generally as simple e in open
unstressed syllables, as relais [rale], petit [pati], reste [resta],
armes [arma], le mattre [I9 meitr], de Paris [da pari]. The
following cases, where it occurs under some other form,
should be noted :
(i) Monsieur is sometimes pronounced with a slack or some-
what indistinct [o] [m6sj0], but the form [m9sj0], or even
[msj0] is more frequently heard.
(2) Peut-etre, though sometimes [p0t£itr] or [pcetsitr], is gener-
ally [p9tsitr].
(3) Soucoupe is often [s9kup] in common speech.
(4) The group ai is sounded [9] in jaisahle [fazabl], faiseur
[f9zoeir], and in the different forms of faire in which ai
is unstressed and followed by s, as faisant [f9za], faisais
[f9z8], faisons [f9z3]. The same is the case with the com-
pounds bienfaisant, bienjaisance, maljaisant, malfaisance.
Under the influence of orthography, however, the sound [e]
is becoming frequent in some of these forms, as [fsza], [fsze].
(5) The pre-fixes des-, -res- are pronounced [d9], [r9] in those
cases where they represent de-, re- (not dS, ri) before s,
as dessous [d9su], dessus [d9sy], ressaisir [r9S8ziir], ressortir
[r9sortiir], ressentir [r9satiir].2 Cresson is also [kr9s3].
The student should guard against the tendency to turn
e mute into close [e] when in an initial syllable. This is an
^ In the seventeenth century the vowel of le, even when stressed, was
frequently elided before another vowel, as in Moliere's line :
Mais, mon petit Monsieur, prenez-l{e) un pen moins haut.
This licence was followed by Victor Hugo and others in later times.
Victor Hugo has even elided e before a full stop in a line in Cromwell :
Chassons-l{e). Arridre, tous I
2 See footnote 2, page 36.
68
MANUAL OF FRENCH PRONUNCIATION
old tendency, which has already affected a large number of
words. For example, crecelle, prevot, pepie, sejour, bent, desert,
peter, petiller, etc., were originally pronounced with mute e.
Desir used to be desir [dasiir] with some writers and the
Comedie Frangaise, but the Academy substituted the close [e]
in 1762. Similarly, rebellion has taken the accent, in spite of
e mute in rebelle and se rebeller ; retable tends to take the place
of retable, through a false analogy no doubt with retablir ;
and celer is taking the place to some extent of celer, under the
influence of receler. Care should, however, be taken to
preserve the mute e in cases where it remains, such as rejuge
(in spite of rejugier), religion (in spite of irreligion), etc. The
difference of meaning between such words as repartir and
repartir, recreer and recreer, reformer and reformer, etc., should
also be remembered.
EXERCISES
Brebis, crevette, frelon, premier,
d6partement, gouvernement,
porteplume
brabi, kravet, fralS, pramje,
departama, guvernama,
portaplym
Je faisais, tu faisais, il faisait,
nous faisions, vous faisiez, ils
faisaient
33 fazE, ty fazE, il fazE, nu
fazjo, vu fazje, il faze
res-
Ressaut, ressauter, ressemblance,
resserrer, ressource, ressouvenir
raso, rasote, rasablais,
rassre, rasurs, rasuvnirr
Peut-etre monsieur veut-il une sou-
coupe.
Je pars demain, le temps reste beau,
Le cour de justice rentrera mercredi.
11 faisait quelquefois des porte -plumes.
Ventrebleu ! Charles ressemble k un
gredin !
pate:tr masj0 votil yn sakup.
33 pa:r dame, la ta resta bo.
la ku:r da 3ystis ratrara
merkradi.
il faze kelkafwa de porta-
plym.
vatrabl0 ! Jarla rasflibl a de
grade !
CHAPTER XVII
THE SEMI-CONSONANTS
71. We have already pointed out (§ 7) that there are three
sounds, represented phonetically by [w, j, q], which are not
merely vowels but consonants also, inasmuch as they are
partly produced by a distinct friction of the breath. They
are therefore called semi-consonants. They are really the
vowels [u, i, y] (at the top of the triangle), produced with still
narrower opening of the lips, and with tongue raised still
higher towards the palate. Referring to the diagrams in
§ 8, it will be observed that for the vowels [u, i, y], the lip-
opening is very small and the passage between the tongue
and palate is quite narrow. Now, if one of them is produced
immediately before another vowel (as in oui, pied, nuit) so as
to make one syllable of both vowels, an extra tenseness is
required which narrows the passages still more. The lip-
opening becomes smaller than for any pure vowel, and the
tongue rises so high as almost to touch the roof. The result
is audible friction, and the sound is really a semi-consonant.
The vowel [u] becomes the semi-consonant [w] (pronounced
like w in English win) ; the vowel [i] becomes [j] (pronounced
like y in English yes) ; and the vowel [y] becomes [q], a sound
not found in English but explained below. Thus we have
oui [wi], moi [mwa], vienne [vjen], Dieu [dj0], suis [sqi], muet
[mqs]. It is evident that these semi-consonants may be
placed at the extreme top of the vowel-triangle (§ 103).
The Semi-Consonant [w]
72. As already stated, this is the same sound as the
English w (in win, with, weak, etc.). All one has to do for
its clear enunciation is to dispose the lips and tongue exactly
as for [u] (as in French sou), but to pass immediately to the
70 MANUAL OF FRENCH PRONUNCIATION
pronunciation of the second vowel. It occurs as follows in
ordinary spelling :
(i) Where ou is followed by a vowel, ^ as ouest [west], ouate
[wat], couenne [kwan]. The semi-consonantal sound is also
found in rapid speech in cases where one word ends in [u]
and the next word begins with a vowel. Thus, ou est-il,
in ordinary slow speech is [uetil], but uttered quickly it
becomes [wetil].
(2) In the groups oi, oi, oy, oie, pronounced [wa], [wa]
(§ 38, 41), as oiseau [wazo], croitre [krwaitr], noyer [nwaje],
joie [3wa].
(3) In the groups oe, oe, as moelle [mwal], moellon [mwalo],
poele [pwail]. See § 38 (7).
(4) In the nasal -oin, as loin [Iwe], moins [mwe], poing [pwe].
(5) As w or wh in words of English origin, as tramway
[tramwe], whist [wist], whig [wig] ; also in wallon [walo].
^ Where ou follows a group formed of a consonant ^r or /, it remains a
full vowel, as : s'ebrouer, dcrouelles, prouesse, fioueur, clouage, etc. The
same is the case in such words as boueux [buo), noueux [nu0], where
a full vowel seems preferable. To these must be added such verbal
forms Qs j onions [3uj3], jouiez [3uje], etc.
EXERCISES
ou
Ouailles, ouaige, ouais, ouir,
ouie, douane, fouace, douai-
riere, enfouir, souhait, jouer,
denouer
wa:j, we:3, we, wi:r, wi,
dwan, fwas, dwerjeir,
afwi:r, swe, 3we, denwe
oi
Oisif, boite, voile, foyer, soie
wazif, bwa:t, vwal, fwaje,
swa
oin
Coin, point, soin, groin, pointe
kw8, pw8, SW8, grwE, pw£:t
w
Sandwich, railway, warrant
sadwitj, rElwe, wara
Je vois le roitelet dans le wigwam.
Antoine s'assoit sur la ouate.
Voici des oiseaux noirs pour toi.
J'aper9ois trois voiles a I'ouest.
Dubois fait-il la moisson ce mois ?
139 vwa la rwatle da la wig-
wam.
atwan saswa syr la wat.
vwasi dez wazo nwa:r pur
twa.
3apErswa trwa vwal a 1 west.
dybwa fst il la mwaso sa
mwa?
CHAPTER XVIII
THE SEMI-CONSONANT [j]
73. This sound, generally named ' yod ' by phoneticians,
is simply the English y in such words as yet, yes, lawyer, etc.
For its proper enunciation, the tongue rises so high in the
front that the breath cannot pass without friction. It is
this friction, added to the vocal effort, that constitutes the
sound. It is always found in ordinary spelling as i, I, or y,
and may occur initially, medially, or finally.
74. (i) When occurring initially, the pronunciation of
* or y is generally [j], as in ionique, iota, iule, etc. In a few
words, however, it hesitates between [i] and [j], and there is
consequent indecision in the number of syllables. Thus,
hiatus is generally pronounced with [i], and consists of three
syllables [i-a-tyis], but often it takes [j] and has only two
[ja-tyis]. The same may be said of hyene, yeuse, and one
or two other words. Even such a common word as hier is
not always sounded with [j]. Thus, one hears often la soiree
d'hier [la sware di-sir], with two syllables, and avant-hier
[avatisir] with four, and this, indeed, has been regarded as
the proper pronunciation of hier since the sixteenth century.
The prefix hier-, however, takes [j], as hier archie [jerarji],
hieroglyphe [jeroglif].
75. (2) When medial : In old French the letter i after an
r ov I was a semi-consonant [j] or a vowel [i] according to the
etymology of the word. Thus, un sanglier, un ouvrier, were
pronounced with [j], but marier, vous liez, with [i]. To-day
etymology is no longer a ruling factor, and the French use
[j] when there is a vowel before r or 1, and [i] when
there is a consonant. Thus, they say with [j], marier
[marje], vous liez [vu Ije], nous voulions [nu vuljo], la liaison
[la Ij8z5] ; but with i, sanglier [saglie], ouvrier [uvrie], vous pliez
[vuplie], nous rdclions [nu raklio], une mauvaise liaison [liezo].
71
72 MANUAL OF FRENCH PRONUNCIATION
In verse, which is regulated in French principally by the
number of syllables, this rule is not always adhered to. Prose,
e.g. makes no difference between nous passions and les
passions, while verse puts [j] in the first [pasjo], and [i] in the
second [pasio]. Similarly, prose puts [j] in such words as
lion, ambition, emotion, epie, odieux [lj5, abisjo, emosjo, epje,
odj0], while verse puts [i] in them and thus lengthens the
number of syllables ([lio], etc.). For such variations from
prose rules, reference should be made to treatises on
versification.
76. The combination -ill is common both in the body of
a word and at the end. When following a vowel, it is simply
pronounced [j], as travailler [travaje], taillis [taji], hdillon
[bajo], paille [paij], feuille [foeij] ; ^ but when following a
consonant it takes the sound [ij], as pillage [pijai3], brillant
[brija], bille [biij]. In the latter case there are some excep-
tions, in which II is pronounced [1] (see § 115).
77. The letter y between two vowels becomes i-i. The
first i combines with the preceding vowel, and the second
is pronounced [j]. Thus, rayon becomes rai-ion, i.e. [rejo] ;
royaume becomes roi-iaume [rwajoim] ; and soyons becomes
soi-ions [swajo].
On the same principle [j] tends to introduce itself between
i and another vowel, as prier [pri-je], nous prions [prijo],
prieur [prijoeir], triage [trijaij], crions [krijo], etc.
78. When final : The termination -il after a vowel is
pronounced [j], as travail [travaij], pareil [pareij], deuil [doeij].
The letter I in linceul takes the same sound as a rule [lesoeij].
In such words, instead of final [j], some people utter some-
thing like [i], short and rapid, or even nothing at all. Thus,
soleil becomes [solsii] or [solei] instead of [solsij]. Others,
again, sound [jo], adding an e mute to the ' yod ' and thus
putting another syllable to the word, which is the other
extreme. These are mistakes and care should be taken to
sound the [j] correctly, without either neglecting it or
exaggerating it.
79. The pronoun y, or i at the end of small unstressed
1 In those cases where -ill is followed by i and another vowel, the
two ' yods ' unite into one, as bailliage [baja:3], joaillier [3waje],
medaillier [medaje].
THE VOWELS
73
words such as qui, si, frequently becomes [j] in rapid speech
before another vowel, as fa y est [saje], Ihomme qui est Id
[kjela], 5^' on veut [sj5v0]. But whenever the y or i becomes
stressed, it retains its pure sound, as qui est Id [kiela] ?
EXERCISES
Init.
lonique, iota, hyacinthe, yeux,
yacht, yole, Yolande
ionik, jota, jase:t, J0, jak,
jol, jola:d
Med.
Encrier, tablier, prendriez, crier,
priere, triage, oublier, etrier
akrie, tablie, prfidrie, krie,
prieir, tria:3, ublie, etrie
Soulier, sciure, liane, petiot,
piano, pion, comptiez, moitie,
bien, j&er
sulje, sjy:r, Ijan, patjo,
pjano, pj3, kotje, mwatje,
bje, fj8:r
Mouiller, veiller, cuiller, veuillez,
piller, billet, sillon, griller
muje, V8je, kqiJ8:r, voeje,
pije, bij8, sij3, grije
Balayer, effrayer, noyer, noyons
bal8je, efreje, nwaje, nwajo
Fin.
Bail, corail, email, reveil, vieil,
conseil, accucil, orgueil, cer-
feuil
ba:j, koraij, ema:j, reve:j,
vj8!j, koseij, akceij, orgoeij,
S8rfoe:j
Assieds-toi pr^s des hyacinthes.
Julien et sa famille sont idiots.
Le chien qui aboyait est a Pierre.
Tiens bien tous ces diamants.
J'ai paye la viande hier.
asjetwa pre de jaseit.
3ylj8 e sa fami:j sot idjo.
io JJ8 ki abwaje et a pje:r.
tje bje tu se djama.
3e peje la vjaid jeir.
CHAPTER XIX
THE SEMI-CONSONANT [q]
80. This is the sound [y] when it comes before another
vowel (it generally comes before i). The lips and tongue
are disposed exactly as for [y], but the vocal apparatus passes
immediately and smartly to the vowel which follows, as
suis [sqi], nuage [ni[ai3], muei [mqe]. The beginner finds
some difficulty in producing this sound, being apt to dwell
too much upon the [y], thus making two syllables, instead
of combining them as one. Let him try to articulate the
[y] rapidly, shortening it as it were, and concentrating his
thought on the next vowel from the beginning. In this way
he will, with persistent practice, insure a good pronunciation
of [q], with both vowels linked into one syllable. The sound
is specially difficult after r, and constant repetition is needed,
as in ruisseau [rqiso], bruit [hTqi], fruit [frqi].^
Referring to the triangle in § 103, it will be seen that [q]
is reaUy the mixed sound between [j] and [w], i.e. it is
[j] pronounced with lips rounded and projected as for [w].
This idea may help the beginner towards its correct enuncia-
tion in some words at least, where [q] can be so pronounced
without much difficulty, e.g. puis, nuit, suite, suivre, fuir, etc.
The main fault with English speakers is to pronounce
[q] carelessly as [w]. Lui becomes [Iwi], identical with
Louis, thus confusing two distinct French sounds. The fault
is very common, and is due to making the first sound [u],
instead of [y] or rounded [j].
81. In cases where the letter y comes between the two
sounds (as in appuyer), it divides as usual (§ 78) into i-i,
* In many cases where the u is preceded by two or more different
consonants, it remains a distinct vowel, as obstruer [opstrye], concluant
[koklya], ftuide [flyid], druide [dryid], etc. But it has become a semi-
consonant in bruit, fruit, autrui, pluie, truie, truite, detruire, instruire,
construire, and a few other cases. In duo it is a full vowel [dyo].
74
THE VOWELS
75
as appui-ier, which is thus pronounced [apqije]. Other
examples are : essuyer [es\j^]e],fuyard [fqijair], tuyau [tipjo].
Some French people, however, are content with one i in
some words of this kind, especially where the sound follows
[r]. Thus, hruyamment becomes simply bru-iamment, i.e.
[bryjama], so that instead of the difficult group [qij] we have
merely [yj]. The same remark applies to hruyant [bryja],
bruyere [bryjeir], gruyere [gryjeir], etc. Even tuyau, tuyere
are frequently [tyjo], [tyjeir].
82. The combinations gu- and qu-, which are pronounced
[gq] and [kq] in some cases, are referred to under the con-
sonants (§§ 157, 161).
EXERCISES
ui
Buis, puits, puissant, tuile, muid,
nuire, bruire, juin, aujourd'hui
biji, pqi, pt[isa, tqil, mqi,
niiiir, brqiir, 31^8,1 osurdqi
hui
Hultre, huile, huit, huissier
qitr, qil, qi, qisje
Nuee, remuer, situe, Su^de, per-
suader
nqe, ramqe, sitqe, sqed,
psrsqade
ni
Ruer, druide, ruisselant, truite
rqe, drqid, rqisla, trqit
uye
Ennuyer, ecuyer, gruyer, Gruyere
anqije, ekqije, gryje, gryjeir
Les ennuyes fuient le bruit.
11 cuit la truite aujourd'hui.
lis ne conduisent pas des truies.
Je suis aiguilleur depuis juin.
Le bruit du niisseau m'ennuie.
lez anqije fqi I9 brqi,
il kqi la trqit 03urdqi.
il na kodqiz pa de trqi.
39 sqi egqijoe:r dapqi 3q£.
la brqi dy rqiso manqi.
^ The pronunciation [3W8] is frequently heard in Paris.
CHAPTER XX
THE NASAL VOWELS
83. Nasal (or nasalized) vowels are produced, as explained
below, by lowering the soft palate, and thus allowing
part of the sound to escape through the nose. In this
way the resonance of the nose is added to that of the mouth.
English has no such vowels, but in French the four vowels
[a, 0, oe, 8] occur nasalized, being represented phonetically
as [a, 5, oe, e].
84. In ordinary spelling they appear as vowels with n or
m attached, but care must be taken not to sound the n or m,
except where liaison is allowable. Thus, tante is [tait], not
[taint]. Similarly, we have oncle [oikl], jardin [3ardE], humble
[deibl], un hon vin blanc [de bo ve bla]. The pronunciation
is exactly the same whether the next letter is n or m ', the
words 7ion and nom, for example, are pronounced absolutely
alike.
85. In regard to length, they are only short when final
in pronunciation, as gant [go]. In all other cases they
are either long or half-long, being long when under stress,
as reponse [repois], and half-long when unstressed, as
embaumer [a'bome]. For most practical purposes, however,
' half -long ' may be regarded as short.
86. It should be noted that there is no nasal sound in the
following cases :
(i) If m or n is followed by a vowel (in the same
word),i as amour [amuir], animer [anime], colline [kolin].
Masculine adjectives ending with a nasal sound thus lose
it in the feminine, which takes a different sound. For
example, un [ce], une [yn] ; gamin [game], gamine [gamin] ;
1 The reason for this is that, owing to the division of syllables, the
morn goes with the following vowel.
76
THE VOWELS 77
flein [pie], pleine [pie in] ; Parisien [parizje], Parisienne
[parizjen]. Other examples :
Followed by a vowel : Ami, image, homogene, synonyme, brume,
promener, plume, Nanine.
Feminines : Voisin, voisine; fin, fine; brun, brune; sain,
saine ; prochain, prochaine ; Italien, Italienne ; ancien,
ancienne ; doyen, doyenne.
(2) If there are two m's or two n's together (in the
same word), as gemmation [3emasjo], annee [ane], etrenne
[etren], homme [om], dilemme [dilem]. Other examples :
Nommer, comment, savamment, pommier, flamme, sonner,
donner, anneau, honneur, renne, colonne, annales, Cinna.
The prefix en- or em- (or rem-) is always nasal, however,
whatever letter follows. A vowel after it, or the doubling
of the n or m, does not change it. Thus, eni:prer, i.e. en-ivrer,
becomes [anivre], the nasal sound being preserved and the
n being carried forward in liaison, as in s'en alter (see § 102) ;
and ennobtir, i.e. en-noblir, becomes [anobliir]. Similarly,
we have enorgueillir [anorgoejiir], emmagasiner [amagazine],
ennui [anqi], remmancher [ramaje], and many others.^
The prefix im- is nasal in the two words immangeable
[8ma3abl], immanquahle [emdkabl].
(3) If m is followed by n (in the same word), as am-
nistie [amnisti], somnambule [somnabyl], automnat [otomnal],
gymnase [3imnaiz]. Other examples :
Omnibus, calomnie, hymne, omnipotence, amnesic, somnolent.
Exceptions are automne [oton], and damner [done] with its
derivatives.
On the other hand, if n is followed by m, the nasal sound
is retained, as in neanmoins [neamwe], ttnmes [teim], vtnmes
[veim].
(4) In many foreign, classical, and unfamiliar words,^
^ The em- is not nasal, however, in such words as Emmanuel, Emmaiis,
enneagone, etc., where it is not the prefix.
* This is specially the case with final m. Only a small number of
such words have the nisal sound {dam, Adam, daim, faim, essaim, etaim,
thym, nom (and compounds), dom, parfum), the rest having lost it since
the seventeenth century. On the other hand, final n {-an, -en, -in {-ain,
-ein, -oin), -on, -un) generally means a nasal sound, except in the
terminations -en after a consonant (mostly foreign words or proper
78 MANUAL OF FRENCH PRONUNCIATION
as Amsterdam [AmstErdam], interim [sterim], museum
[myzeom], specimen [spesimen], decemvir at [dessmvira]. Other
examples :
Abraham, harem, item, Jerusalem, Kremlin [kremle], maximum,
opium, Eden, lichen [liksn], gluten, dolmen, Siam, Bethleem,
abdomen, Potsdam, macadam, rams [rams].
THE PRONUNCIATION OF THE NASALS
87. This matter is important, as nasals occur so frequently
in French, and there is nothing to correspond to them in
English. Sometimes there are sentences in which almost
every vowel is a nasal, as Uenfant mange son pain sans
mecontentement. So many false and absurd directions, too,
on the subject are given in some books, that much misunder-
standing exists. Some learners are taught to pronounce these
vowels incorrectly as ordinary oral ones with the English -ng
added to them. The pronunciation will be found to be sim-
plicity itself, if the following explanation be grasped :
Looking at the adjoining diagram, it will be noticed that
the expiratory breath, on leaving the throat, has a double
passage before it : it may pass either through the nose or
the mouth. These two chambers are separated by a hori-
zontal partition, called the palate, which forms the floor
of the one and the roof of the other. The front portion of
this partition is a bony, rigid structure,' known as the hard
nouns), -man (particularly in English words), -in (in several foreign
words or names), German names in -ein (except Mein), a few learned and
foreign words in -on, and most English proper names in -son and -ton.
Thus the sound is not nasal in the following and many others, in
addition to examples given above :
Amen, albumen, cerumen, cyclamen, gramen, rumen, pollen,
hymen {except in rhyme sometimes), Aden, Baden, Carmen, Baylen,
Dryden, Ibsen, Niemen, Yemen; alderman, clubman, gentleman,
recordman, yeoman, policeman, sportsman, Ahriman, Flaxman,
Wiseman, Wouverman ; djinn. Khamsin, muezin, Lohengrin {in
music at least), gin {sometimes), Darwin, Erin, Elgin, Erwin, Stettin,
Emin-Pacha, Robin Hood ; Holbein, Gerolstein, Rubinstein, Zoll-
verein ; epsilon, omicron, sine qua non, megaron, baralipton,
singleton, Byron, Lang-son, Satyricon ; Addison, Emerson, Hudson,
Nelson, Tennyson, Fulton, Hamilton, Palmerston, Washington,
Wellington, etc.
Of all the words in -en after a consonant, of French or foreign origin,
ex amen is practically the only one that has the nasal sound [egzamg].
THE VOWELS 79
palate (H). The back portion is composed of muscular
tissue, soft and flexible, and is known as the soft palate
or velum (S). This soft palate terminates in a conical body,
Fig. I
called the uvula, which hangs suspended from it. Now, it
is evident from the diagram that the soft palate, with its
attached uvula, forms a kind of curtain between the mouth
and nose. Being very flexible, it can be raised or lowered
at will, and the passage into the nose can be closed or opened
accordingly. There are three possibilities indeed :
(i) The velum may be lowered right down towards the
tongue. In this case the air is prevented from entering the
mouth and is sent through the nasal passage. This is the
way taken by the breath in ordinary respiration when we
shut the mouth and breathe through the nose : the velum
descends of its own accord, and the air passes to the back
of it. It is also the way taken in the pronunciation of m, n,
and English -ng, for in m the air is stopped by the closing
of the lips from passing out of the mouth, in n by the tip of
the tongue on the hard palate, and in -wg by the back of the
8o MANUAL OF FRENCH PRONUNCIATION
tongue rising up to meet the soft palate, the air in each case
being directed into the nose. It is thus possible to produce
sound in this way, but in such cases we have a nasal one
alone.
(2) The velum may be raised and pressed back against
the pharynx wall (PP). In this way the passage into the
nose is closed, and the air is forced to pass through the mouth.
This is the position of the soft palate during mouth breathing,
and also during the articulation of all English vowels and
all consonants except the three just mentioned. This may
easily be verified. Let the student hold a mirror before his
mouth in a good light, so as to see well into the inside. Let
him take in breath through the nose with the mouth open,
and then let it out forcibly through the mouth. When taking
the breath in, it wiU be noticed that the velum descends
towards the tongue, while in giving the breath out, if done
sharply, the velum will be seen to rise.
(3) The velum may be lowered sufficiently to allow
both passages to be utilized at the same time. In this
case the way into the nose and that into the mouth are both
free, some of the air being deflected into each. The resonance
of the nose is thus added to that of the mouth, so that we have
nasal resonance together with the ordinary sound, and this
is what takes place in the production of the French
nasal vowels. The velum is not completely lowered, but
sufficiently to allow part of the sound into the nasal cavity.
It is evident that any vowel can thus be nasalized, but in
French only the four already mentioned are thus affected.
The sounds [a] and [5] require good nasal resonance, while
[de] and [e] require less. In the case of purely oral sounds,
the velum may sometimes fall a trifle, and a little air passes
into the nose. This is what frequently happens in Portuguese
and in American English. Some Americans indeed produce
a definite nasal twang which does not belong to pure English.
A slight increase of ordinary nasal resonance is all that is
required for the two latter sounds.
88. There are some methods of developing the nasal vowel
sound in the case of those learners who find it difficult :
(i) Humming should be tried, for here the velum is
lowered and the air passes through the nose. Sustain a
humming tone for a couple of seconds ; then all at once, without
THE VOWELS 8i
stopping the tone or moving the velum, open the mouth and
sound [a]. It has the same effect as singing very softly the
syllable [ma], the nasal vibration being distinctly felt. After-
wards, on the same pitch hum [n], and again sing very softly
the syllable [na]. Practise this exercise on various tones,
changing occasionally to the vowel [5].
(2) The changing of [b] into [m] lowers the velum.
For [b], the velum is raised high, and touches the back of the
pharynx wall, and all the breath employed thus escapes by
the mouth only. For [m], the position of the lips and tongue
is approximately the same, but the velum descends, so that
the breath passes through the nose only. The same is the
case with [d] and [n]. In the production of a nasal vowel,
therefore, all that is needed is the same treatment that changes
[b] into [m], or [d] into [n].
In practising the nasal sounds, the sensation connected
with the fall of the velum should be carefully noted, so that
it may be produced at will. No effort is needed to lower
the velum : it is done almost or quite unconsciously when
the sensation of its movement becomes familiar.
89. A common fault with beginners is to sound the vowel
as a pure oral one, and then add the English nasal consonant
-ng (phonetic sign [r)]) to it. Thus, we have such ridiculous
pronunciations as ' ahngfahng ' [agfar)] for enfant, * tahngt '
[tagt] for tante, ' bong ' [borj] for hon. This fault is largely
due to misleading directions in grammars and other books,
whose writers fix upon -ng as the nearest English nasal sound,
hoping that it will give some idea of the pronunciation. For
the production of -ng the tongue rises up at the back towards
the velum, thus shutting off the mouth and sending the sound
into the nose. It is therefore a nasal consonant, like m and n,
but it is practically unknown in the French language, and
its addition to a vowel does not make that vowel a nasal one.
What happens is that we have two consecutive sounds, viz.
a pure oral vowel followed by a pure nasal consonant. In
other words, the current of sound commences to pass solely
through the mouth, and then finishes by passing solely through
the nose. In some books the learner is counselled not to
pronounce the -ng fully, but this direction does not alter the
matter, as we still have a pure oral vowel with the semblance
of a nasal consonant at the finish. It should be clearly
82 MANUAL OF FRENCH PRONUNCIATION
understood that the French nasals cannot be properly produced
in this way : they require nasahty throughout their whole
duration, and not added at the end. The nasality com-
mences with the first vibration of the vowel and finishes
with the last.
90. Another fault, even with some Frenchmen, is to sound
the n or m which follows in the ordinary spelling. This was
the correct pronunciation in olden times, and it still lurks
in certain dialects, especially in the south of France ; but not
the slightest trace of these consonants should now appear
in the pronunciation so far as correct Parisian or northern
French is concerned. The fault is specially noticeable in the
following cases :
(i) When n comes before d or t, as in ianfe, plante, viande.
The tendency to sound the n here is due to the fact that n
takes approximately the same tongue position as <^ or ^ (viz.
tip of tongue against front of palate), the only difference
being that n requires the velum down. In getting ready,
therefore, to sound the ^ or ^ while the velum is still lowered
for the nasal vowel, there is an easy tendency to sound the n.
(2) When m comes before b or p, as in lampe, ombre,
jamhon. Here the same cause operates, for m requires the
closing of the lips, and this is precisely the position for h or p,
the only difference being that in m the velum is lowered. In
thinking, therefore, of 6 or ^ too soon, while the velum is still
down for the nasal vowel, there is a tendency to sound the m.
(3) When n comes before c (k) or g, as in encore, enclos,
anglais. Here the n is apt to become -ng, as this takes the
same tongue position as k or g, with the difference that it has
the velum down.
The remedy in all these cases is to finish the nasal vowel
before beginning to pronounce the following consonant, i.e.
the velum must be raised before the consonant is sounded.
As a help such words should be practised in separate syllables
as van-ter, sen-tir, jam-bon, en-core.
The Nasal [a]
91. This vowel is the nasalized form of [a], i.e. the mouth
is wide open as for the ordinary vowel, but the velum hangs
down to permit the entrance of sound into the nose. Many
THE VOWELS 83
French speakers put a touch of [o] into the sound by making
the point of articulation intermediate between [a] and [o].
In emphatic or affected speech, indeed, and frequently among
children, [a] may tend to become full [5] .
It is represented in ordinary spelling by an, am, en, em,
aon, aen (or aen), as plan [plci], plante [plait], champ [fa],
gens [3a], temps [ta],faon [fa], Caen [ka]. In one or two cases,
however, aon is pronounced [ao], as Lycaon [likao], Pharaon
[farao] ; and before [n] it becomes [a], as faonner [fane],
paonne [pan], paonneau [pano], Laonnais [lans], Craonne
[kran]. In Jean the e is mute [3a].
92. Owing to the variety of spelling referred to, many words
with [a] are pronounced the same, though written differently
(Homonyms) . Thus we have :
Sang, sans, sens, sent, cent, e'en, s'en (all pronounced [sd]).
Tan, tant, taon, tend, tends, temps, t'en (all pronounced [ta]).
The Nasal [5]
93. This is not exactly the nasalized form of [o], but rather
of middle [0], that is, a vowel intermediate between [o] and
[o] (see § 51 (2)). Indeed, some phoneticians would prefer
to write it [6]. But all that is needed is to round and pro-
ject the lips a little more than for [o], while at the same
time lowering the velum. Careful practice of this will soon
give the correct sound.
94. The two vowels [a] and [5] should be clearly dis-
tinguished from each other. Many beginners fail to do this :
they pronounce both of them incorrectly, replacing them by
a fully open [o] nasalized. For [a] the tongue lies flat, and
the mouth is wide open, while for [5] the tongue rises a little
to the back, and the lips are closer and rounded. If these
characteristics are remembered, there should be no confusion.
95. The nasal [5] is represented in the current spelhng by
on, om; and also by un, um in foreign or borrowed words,
as conte [koit], nom [no], lumbago [lobago], jungle [30igl],
junie [30 it].
The Nasal [e]
96. The sound [e] is really [se] nasalized, and ought to have
a special symbol (not [§]), but as there is no advantage in
84 MANUAL OF FRENCH PRONUNCIATION
having too many symbols, [e] is used by phoneticians. The
[ae] is the Southern English or American (not Scotch)
sound of a in such words as man, mad, rat, etc. Those who
have heard the query, " Cab, sir ? " in an English town will
remember the sound of the a. It is neither [c] nor [a], but
intermediate between. It is almost unknown in Scotland,
but is the common sound of close [a] in most parts of south
England and of America. The sound [e] is exactly the nasalized
form of this English [se]. Some students find it diihcult
to produce it, and are inclined to sound either pure [e] or
go to the other extreme and sound [a], but if they will take
advantage of the many opportunities of picking up the sound
of the English or American [^e], all they have to do is to give
it a little tenseness and nasalize it.
97. It occurs in spelling in a great variety of forms —
wherever there is an i or y followed by w or m. Therefore in,
im, yn, ym, ain, aim, ein, eim, are all pronounced [c], as fin
[fs], impot [epo], synthase [seteiz], nymphe [neif], main [me],
faim [fe], plein [pie], Rheims [reis]. Similarly, the syllable
-oin becomes [we], as loin [Iwe], poing [pwe], coin [kwe],
poinie [pweit].
98. The variety of spellings, as in the case of [a], gives rise
to numerous homonyms, as :
Sain, saint, sein, seing, ceins, ceint, cinq (all pronounced [se]).
Vin, vins, vint, vingt, vain, vainc, vaincs {all pronounced [ve]).
Tin, tins, tint, tain, teins, taint, thym {all pronounced [te]).
99. The nasal sound -en at the end of a syllable is pro-
nounced [e] (not [a]) in the following cases :
(i) In the terminations -ien, -yen, -een, as hien [bje], cJiien
[Jje], pa'ien [paje], citoyen [sitwaje], chaldien [kaldee].
Derivatives follow the same rule, as hienjaisant [bjef9za],
hientot [bjeto], etc.
(2) In certain tenses of venir and tenir, with their deriva-
tives, as viens [vje], viendrai [vjedre], tiens [tje], maintien
[metje],
(3) In many foreign words/ as appendice [apedis], benzine
[bezin], examen [egzame], menthol [metol], peniateuque
^ The reason for this is that the syllable -en, in becoming ' Frenchified,'
can only do so directly by [e], the sole nasal corresponding to e.
THE VOWELS 85
[pEtat0ik], and all other words from Greek TrsvTt (except
pentecote [pdtkoit] Also in Saini-Ouen [setwe], and many
proper names, especially in -ens where the s is sounded, as
Martens, Camoens, Saint-Gaudens, etc.
In all other cases the syllable -en preserves the [a] sound,
as client [klici], orient [orja], expedient [ekspedja], audience
[ocljais].
The Nasal [de]
100. The oral vowel [oe] is [e] with the lips rounded, but
the nasal [de] is a little more open : it is really the nasal
[se], i.e. [e], with the lips rounded, and may thus be placed
in the middle of the triangle opposite [se] (see triangle, § 102).
Properly speaking, it should not be represented phonetically
by the sign [de], but phoneticians have adopted this sign
as representing the nearest French sound. To produce it
correctly, all that is needed is either to nasalize [oe], making
it at the same time a little more open, or to sound the
nasal [e] with lips a little rounded and pushed out. It is not
a sound of frequent occurrence in French, apart from the
article un. In ordinary spelling it is represented by un or
um, as hrun [brde], parfum [parfde]. It also occurs as eun
in a jeun [a 306], and in Meung [mde].
101. Liaison. — Sometimes a final n (never m), indicating
nasality, is carried forward in liaison, i.e. it is pronounced
before the next word if this begins with a vowel or h mute
(see Chapter XXXIII). But it is so sounded only in ad-
jectives followed by a noun (e.g. commun accord) ; in mon,
ton, son ; and in en, on, un, rien, bien (and sometimes combien),
when there is a direct connection between these words
and the next, as in il en a, en Italic, on apprend [ilana, anitali,
onapra].^ In such cases the vowel occasionally loses a little
of its nasality. The vowel -ain [e] in adjectives is the most
subject to this phenomenon, and frequently loses all the
nasality, the sound heard being that of the feminine ([e]),
as certain auteur [ssrtEU otoeir], un mien ami [de mJEU ami].
Similarly, we have un vain espoir, un ancien usage, un vilain
enfant, en plein air, le moyen age, au prochain avertissement,
^ There is no liaison, therefore, in such expressions as vain et faux,
ancien et demode, bon a rien, bon a tirer, un ou deux, un et un font deux,
run est venu, parlez-en a voire pere, a-t-on et^, rien ou peu de chose, nous
sommes bien ici, bien et vite, etc.
86 MANUAL OF FRENCH PRONUNCIATION
etc. The vowel [5] loses all its nasality in bon, as un bon
eleve [de bon eleiv]. Compare bonheur, bonkomme. In the
case of words ending in -in, denasalization only occurs with
divin, particularly in the expression divin enfant, the sound
here being also that of the feminine [divin dfa]. It is found
frequently, too, in such phrases as divin Achille, divin Ulysse,
divin Homere, etc. But in all other cases the nasal sound is
retained in the liaison, as malin esprit [malen Espri], fin esprit
[fen espri], etc. In the case of words in -un, denasalization,
once common, is now unknown, and we consequently have un
homme [oen om], un ami [den ami], un un [cen ce], un a un
[den a de], Vun et V autre [Iden e loitr], aucun homme [okden om],
chacun un [Jakden de]. The words en, on, rien, bien, combien
are never denasalized, doubtless owing to the fact that they
cannot have a feminine, as je n'en ai pas [39 ndn e pa], on a
dit [on a di], rien a dire [rjen a diir], bien aimable [bjen emabl],
combien avez-vous de . . .? [kobjgn avevu da . . . ?].
The modern tendency is to denasalize as little as possible,
owing no doubt to the perpetual confusions that would result
between the masculine and feminine forms, although many
outstanding teachers prefer denasalization as more in line with
the development of the language. Such words as an-nee,
solen-nel, ardem-ment, etc., were once nasal.
102. The vowel triangle may now be fully constructed as
follows. Sounds not occurring in English are marked with an
asterisk.
(Semi-consonant) f 4.* (Semi-consonont;w
(Clo^)l y", _ (Clos^u
COp^n)i (Cbeh)a
(Cl>se)e ^* ,(Cljk)o
(Middle) e (Middle; 0,
(^p^t CZ* (Open}0
(More open; 2B,l^~C£* .(More6per:)c(atin"aubi'rwol'J
(Obsc) a
(Twiddle) ^
(C^a.a'
THE VOWELS
S7
EXERCISES
em-
en-
Emmailloter, emmancher, em-
menager
Enamourer, ennuyer, enhardir
amajote, amaje, amenase
anamure, anqije, aardiir
-nt-
-nd-
Chanter, conte, dente, lenteur,
aimante, teinte, saintete,
fonte
Grandeur, ronde, peindre, con-
traindre
Jate, k5:t, date, latoeir,
emate, teit, sstte, fo:t
gradceir, r3:d, psidr, kotreidr
-mb-
-mp-
Embl^me, nombreux, plom-
beux, ombre
Impossible, empreindre, lam-
piste, un peu
ablEim, n3br0, pl3b0, o:br
eposibl, apreidr, lapist, oe p0
-nc-
-ng-
Bancal, encre, vaincre, encourir
Sangiot, anglican, engloutir,
engager
bakal, a:kr, vEikr, akuriir
saglo, aglika, agluti:r, agase
an
am
en
em
aon\
aenj
Banc, quand, blanche, change,
rendre
Tambour, camp, rampe, Gam-
betta, Adam
Henri, Rouen, mentir, parent,
tente
Temps, temple, empire, Luxem-
bourg
Paon, taon, Laon, Craon, Saint-
Saens, Jordaens
ba, ka, blaij, Ja:3, raidr
tabuir, ka, ra:p, gabeta, ada
ari, rwa, matiir, para, ta:t
ta, taipl, api;r, lyksabuir
pa, ta, la, kra, sesais, 3orda:s
on
om
um\
un j
Ton, vont, long, onction, bonte
Nom, plomb, pompe, combler,
prompt
Cumberland, Humboldt, punch,
secundo
to, v3, 13, 3ksj3, bote
n3, pl3, p3:p, koble, pro
k3b£ria:d, 3bol, p3:J, s9g3do
in
im
yn
ym
ain
aim
ein
Vin, vinmes, grincer, instinct,
sphinx
Impur, timbre, simple, Edim-
bourg
Lynx, syntaxe, syndic, syncope
Cymbale, Olympe, thym, sym-
pathie
Bain, maint, vaincs, contraindre
Daim, essaim, Paimboeuf
Sein, peintre, feinte, rein, teint
V8, veim, grese, sste, sfeiks
epyir, teibr, ss:pl, edgbuir
l£:ks, sgtaks, ssdik, sekop
sebal, ole:p, te, sepati
hi, me, vg, k3tre:dr
de, 8S8, psbcef
S8, pe:tr, fs:t, rg, tg
oin
Joindre, pointe, poindre, lointain
3wg:dr, pwgit, pweidr, Iwetg
MANUAL OF FRENCH PRONUNCIATION
ien
etc.
Lien, vaurien, Julien, bohemien,
paien,moyen, doyen, europeen
IJE, vorJE, 3ylje, boemje,
paje, mwajs, dwaje, oeropce
For-
eign
Benjoin, vendetta, crescendo,
blende, spencer, agenda,
memento, pensum, Benjamin,
benzine, mentor, Rubens,
Bengale
be3W8, vedeta, kresedo,
bleid, sp£ss:r, a3Eda,
memeto, pesom, bssame,
bEzin, mstoir, rybgis, bggal
un
um
Lundi, aucun, alun, tribun,
defunt
Humble, Humbert, emprunt,
emprunter
loedi, okoe, al&, triboe, defde
de:bl, oebEir, aprde, aprdete
Le commandant entre dans le camp.
Jean plante sa tente devant le banc.
Ton oncle chante la seconde chanson.
Nous allons compter nos compagnons.
Cinq gamins vinrent joindre 1 essaim.
Le nain a besoin de pain : il a faim.
Chacun emprunte humblement a quel-
qu'un.
Quelques-uns des tribuns sont k jeun.
l9 komada a:tr9 da la ka.
3a pla:t sa ta:t dava la ba.
ton 5:kl9 Ja:t la S9g5:d Jaso.
nuz al3 kote no kopapo.
S£ gamg vE:r 3W£:dr9 Iese.
I9 UE a b9zwE d9 pe : il a ie.
Jakde aproeit oebbma a kelkde.
kElkazce de triboe sot a 306.
Quand le Hun en entend le son, il craint.
Le gar9on est tombe jusqu'au fond.
Ton enfant a bon nombre de dents.
Le parent vient d'entrer dans la maison.
Jean de Meung, 6crivain fran9ais, est
d6funt.
Ma tante lave son linge dans la fontaine.
Elle emmene un enfant de quinze ans.
Pendant longtemps nous manquons de
savon.
Le chien mange entierement la viande.
Les olympiens sont importuns.
ka l9 de an ata I9 s3, il kre.
I9 gars 3 £ tobe 3ysko fo.
ton afa a bo n3:br9 d9 da.
I9 para vjg d atre da la mezo.
3a d9 mde, ekrive frasE, e
defde.
ma ta:t la:v so l£:3 da la fotEU.
El amEU den afa d9 ks:z a.
pada lota nu mako d9 savo.
la Jjg ma:3 atJErma la vja:d.
lez olgpJE sot gportde.
CHAPTER XXI
DURATION OF SOUNDS, OR QUANTITY
103. All sounds may be continued for a longer or shorter
time, independent of the rate of speaking as a whole, and
without altering their timbre or nature. This property of
sounds is known as Duration or Quantity. In this chapter
we concern ourselves only with the vowels, and divide them
generally into short and long", with special references to those
that may require half-length. By a short vowel we mean
one pronounced in the shortest time possible, practically
instantaneously, while a long vowel is one ' drawn out ' to
some extent and thus occupying an appreciable time. In
both cases, however, the actual nature of the vowel remains
the same.
104. It is difficult to lay down hard and fast rules on this
subject, as the length of the vowels depends frequently on
the context, on the emotions of the speaker, on the stress
caused by the tonic accent or by emphasis, and on many
other conditions. The following rules, which cover the
majority of cases, will be found sufficient :
(i) The close [e], and the mute [3] are usually short.
Thus etc [ete], les [le], le nez [b ne], peser [paze], rep He [repete],
all contain short vowels. The only occasions when [e] and
[9] become long are when they receive the Emphatic Stress
(§ 215), in which case mute e becomes [oe]. Thus mechantf
[meija], je [306] te nourrirai. The pronoun le also becomes
lengthened when it receives the Tonic Accent, which it does
immediately after an imperative, as je le dis [39ldi], but dis-le
[diloe].
(2) All final vowels are short, even though circumflexed
in ordinary spelling. By a final vowel is meant one that
90 MANUAL OF FRENCH PRONUNCIATION
terminates the pronunciation of a word. There may be a
written consonant after it, but if this is not sounded, the
vowel is spoken of as final, and is short. Thus, the vowel
is short in lit [li], tres [trs], has [ba], trop [tro], tot [to], bout
[bu], genet [39ns], mat [ma] ; in circumflexed verbal end-
ings, as aimdt [sma], 'fit [fi], regut [rasy] ; and similarly
in the final syllables of the following : beaucoup [boku],
objet [ob38], frangais [frase], resolu [rezoly], chapeau [fapo],
profond [prof 5].
As English final vowels are generally long and thus differ
from French ones, the student should be on his guard. The
English words lea and toe, for example, are by no means the
same as the French lit and tot, however much the latter may
be stressed.
If the vowel is followed by e mute, to signify the feminine,
sometimes it is given a slight lengthening (denoted phonetic-
ally by one dot), as venue [vdny], fermee [fermc], amie [ami'].
This is done in certain districts, such as Switzerland, and a
little everywhere in emphatic declamation or sentimental
utterance (e.g. ma hien-aimee), but beyond this the addition
of e mute makes practically no difference in the length of
final vowels. Thus, venu and venue, alle and allee, donne
and donnee are pronounced the same. There may be a slight
shade of difference in stress, the feminines having the accent
more firmly on the final syllable, but in regard to Duration
there is practically no distinction in ordinary speech. If
the French desire to draw special attention to the feminine
form of the word, they prefer to sound the e mute, as is
generally done in poetry, rather than lengthen the final vowel.
Thus, venue in such a case becomes [v9ny9] or even [v9nyoe],
rather than [vany].
(3) The vowels [a, o, 0], and the four nasals are natur-
ally long". It is only when final that they are pronounced
short (in accordance with the rule just mentioned), as degdt
[dega], rot [ro], bleu [bl0], gargon [garso]. Whenever they
are non-final, that is, whenever they do not terminate the
pronunciation of the word, they are as a rule either long or
half-long, being long when stressed (either under the tonic
accent or under emphasis), and half-long when unstressed.^
^ The vowel [a] perhaps conforms less to this rule than [o], [0], and
the nasals.
THE VOWELS
91
This is the case whether they are circumflexed or not. Hence
these vowels are long (assuming them to be stressed) in pate
[pait], tome [toim], meute [m0it], jeune [30in], danse [ddis],
feinte [fgit], humble (deibl], blonde [bloid] ; while they are
half-long (assuming them to be unstressed) in pate [pa'te],
roser [ro'ze], jeuner [30 'ne], danseur [dd'soeir], peinture
[pe'tyir], lundi [Ide'di], longueur [15'goeir].^
There are exceptions in words containing the group [-rwa-].
Some of these have the vowel [a] short, even though it is
non-final. Thus, droite, adroite, etroite, froisse, froide, wide,
paroisse have all short [a]. But apart from these exceptions,
the group [-rwa-] maintains the [a] long, as croire [krwair],
croise [krwaiz], crottre [krwaitr], etc.
Exercise. — In the following table the vowels [a, o, 0],
and the nasals are arranged in groups of short, long, and
half -long. They should be read from left to right along
the lines, and practised carefully till the right length is
acquired.
Short
Long
Half-long
Short 1 Long
Half-long
Bas
base
baser
Mat 1 sabre
jadis
Tas
taille
tailleur
I Trepas Jeanne
gagner
Ramas
ramasse
ramasser
1 Appas 1 Jacque
ma9on
Clos
close
cloture
1 Idiot
pole
poser
Gros
grosse
grossier
Veau
cause
beaute
Chaud
chaude
chaudiere
1 Anneau
saule
pauvrete
Heureux
heureuse
heureusement
1 Deux
neutre
creuser
Odieux
odieuse
odieusement
i Gueux
feutre
deuxieme
Meut
meute
meunier
Queue
creuse
jeudi
Blanc
blanche
blanchir
i Jean
ancre
embl^me
Chant
chante
chanteur
iDent
etendre
sangler
Tant
tante
tantot
Argent
temple
entier
Son
songe
songer
' Done
monstre
rondeur
Pont
ponte
pontife
Fa9on
eponge
allonger
Long
longue
longueur
Maison
Londres
dompter
Vainc
vaincre
vaincu
' Main
timbre
saintete
Plaint
plainte
plaintif
; Vin
mince
teinter
Craint
craindre
craintif
\ Atteint
teindre
quintal
Hun
humble
humblement
i Parfum
Belsunce
Humbert
Empnint
emprunte
emprunter
Verdun
Sund
lundi
^ Half-length is not marked as a rule in this manual, but is included
in ' short.'
92 MANUAL OF FRENCH PRONUNCIATION
(4) In regard to the remaining seven vowels [i, e, a, 0,
u, y,. oe], note the following rules :
I. When they receive the Tonic Accent, that is, when
they are in the final syllable of a stress group, they are long
if followed by a lengthening consonant. The lengthening
consonants are the voiced fricatives^ [v, z, 3, j] and final [r]
(which means an r that terminates the pronunciation of a word).
Thus, these vowels are long in cave [kaiv], pensive [pasiiv],
cerise [s9riiz], trapeze [trapeiz], plage [plai3], college [kol8i3],
soleil [solsij], feuille [foeij], gloire [glwair], vert [veir], aime-je
[8m8i3].
Under this section, however, the following qualifications
require to be added :
(a) These vowels are not only long before final [r], but
sometimes also (especially in poetry) ^ when the [r] is pre-
ceded by a voiced plosive [b, d, g], as in negre, aigre_,
vinaigre, maigre, pegre, cedre, Phedre, etc. They are almost
always long in the termination -evre, as levre, orfevre, plevre,
Sevres, Lefevre, etc. But if the r is followed by another
sounded consonant, even a lengthening one, the vowel becomes,
short, as in cirque, myrte, term, alerte, meurtre, absurde,
marge, carte, charge, larve, serve, marbre, porte, lourde, ours
[urs], etc.
(b) In Paris at least, if the vowel is followed by [b, d,
or g] alone, without [r], it is frequently sounded a little
long, as gUbe [glsib], plebe [plsib], robe [roib], dogue [doig],
aide [eid], il plaide [pleid]. The reason is that the resonance
which precedes the sound of a voiced consonant retards the
explosion of it and renders the vowel a little long. Perhaps
* half-length ' best describes these cases.
(c) The vowel is long in tous [tuis], and veule [voe:l],
these two words being distinguished in pronunciation from
tousse [tus], and veulent [voel]. The words coule, route, ecroule
1 See § 107.
^ There should, of course, be only one pronunciation both in prose
and poetry, but the difficulty sometimes of finding rhymes leads to a
poetic licence ; e.g. finals with a grave accent are frequently rhymed
with those having a circumflex, as creche with preche, centieme with
Boheme, penetre with fenetre, etc. This is largely due to the fact that
long or circumflexed finals are not so numerous as the others. In
ordinary conversation most people would pronounce -ehre, -edre, -egre,
etc., fairly short.
THE VOWELS 93
are also regarded as having a long [u] in careful speech,
but in a familiar phrase like ga route bien, where the tonic
accent is on bien, the vowel [u] is not any longer than in
route.
2. When these vowels receive the Tonic Accent, but
are followed by a consonant other than a lengthening
one, they are generally short, as time [lim], sate [sal], renne
[rsn], sotte [sot], bouche [buj], minute [minyt], neuf [nc^i].
But the following exceptions of importance should be
noted :
(a) These vowels, if circumflexed, are, generally speaking,
long, as tete, bete, chaine, maitre, dottre, abmie, etc.^ At
the same time this is by no means a fixed rule, as there
are numerous examples of the opposite. Thus, the circum-
flexed vowel is generally short in etes (especially when un-
stressed), arrete, pele-mele, gUe, Ue, epUre, dime, dine, flute,
buche, croute, voute, as well as in nous eumes, vous eutes, and
verbal endings {-dmes, -dtes, -imes, -ites, etc.).
(b) When one of these vowels precedes a sounded s
in the ordinary spelling (mostly Greek and Latin words),
it is generally pronounced long, as Cortes [korteis], iris
[iriis], Minos [minois], Brutus [brytyis]. At the same time,
many educated people pronounce it short in such cases ;
and it is generally short in jadis, cassis, mais, vis (a screw),
sus, detritus, plus (in those cases where s is sounded), burnous,
OS, es, etc.
(c) The vowel [s] is nearly always long in the termina-
tions -aisse, -aime, -erne, -ene, and sometimes in the
terminations -aine, eine, as baisse [bsis], gr aisse [grsis],
j'aime [3£im], theme [tsim], scene [sem], haine [em], reine
[rem]. But exceptions are common, especially in the -aine
group ; and altogether the vowel [e] before a non-lengthening
consonant shows such considerable variation that few definite
rules can be laid down. Some speakers make it long, others
short in the same words. The principal thing is the timbre
or essential quality, which remains invariably open.
1 This is due to the fact that the circumflex accent almost always
takes the place of a letter now discarded, generally an s, the presence
of which lengthened the vowel. Thus, tete used to be spelt teste,
fete wsLsfeste, etc.
94
MANUAL OF FRENCH PRONUNCIATION
In some cases of [e], the Quantity is a means of distinguish-
ing words from each other, as, for example :
Long
Short
Long
Short
Reine
La greffe
L'aide
Maitre
lis s'aiment
renne
le greffe
laide
mettre
ils s^ment
Ils s'aident
lis paraissent
Scene, Seine
Veine
L'aine
ils cedent
la paresse j
saine (also long) i
vaine ( „ „ )
laine ( „ „ )
Exercise. — ^The tables below contain the seven vowels re-
ferred to, arranged as long and short. They should be practised
both down the columns and along the lines, until the proper
duration is acquired. Such practical vowel-exercises will not
only ensure ease of production, but will enrich and amplify
the tone :
Long Vowels
i:
e:
a:
o:
u:
y:
Before
V
Rive
616ve
dtrave
ove
prouve
etuve
,,
Olive
glaive
resolve
innove
Louvre
cuve
z
Bise
chaise
toise
Booz
blouse
buse
M
Sottise
treize
moise
Berlioz
douze
ruse
3
Tige
neige
gage
loge
rouge
deluge
^
ObUge
pi6ge
page
toge
bouge
refuge
J
Fille
bouteille
m^daille
oil [o:j]
rouille
l»
Famine
abeille
corail
oille [o:j]
brouille
r
Mourir
Clair
quart
mort
jour
pure
"
Elixir
guerre
poire
encore
court
voiture
oe:
fleuve
^preuve
cueille
feuille
fleur
coeur
Gratis, oasis, Clovis, bis (' twice '), Thalds, Ceres, alods, albatros, Argos, albinos,
Paros, V^nus, Phebus, typhus, blocus, obus, r6bus, etc.
Caisse, baisse, graisse, ^paisse, paraisse, laisse, abaisse, connaisse, paisse, etc.
Also Grdce, Bo^ce, Lucrece, etc. (in sustained speech), cesse, presse, etc.
-toe
Thdme, crfime, systSme, probl^me, emblSme, blaspheme, diad^me, th6ordme,
Nicod^me, etc.
-dne
Scene, ar^ne, sir^ne, ^b^ne, oxyg^ne, phenomene, obscene, hy^ne, indigene,
Athdnes, M6c6ne, etc.
-ame
-eine
Haine, gaine, aine, domaine, marraine, Maine, Lorraine, Touraine, etc.
Veine, reine, Seine, pleine, etc.
6te,etc.
Poete, proph^te, Crete ; zele, hdle, st61e, Philom^le
THE VOWELS
Short Vowels
95
Before
1
Villa
Sublime
Origine
Signe
Pipe
Scribe
Dite
Rapide
Public
Figue
Vif
Riche
Caprice
eternelle
gemme
ancienne
peigne
steppe
Hdbre
dette
Su6de
grec
b^gue
bref
s^che
espdce
! a
0
sandale
parole
dame
Rome j
cabane
couronne |
compagne
rogne
etape
salope
syllabe
robe
date
pelote
aubade
commode t
sac
bloc 1
vague
drogue j
carafe
etoffe 1
vache
reproche
glace
noce
poule
goum
Rangoun
groupe
double
goutte
coude
bouc
fougue
bouffe
farouche
secousse
j mule
1 ecume
1 hme
I repugne
! jupe
i tube
I chute
I rude
] due
: conjugue
! tuf
j autruche
I puce
gueule
peuple
meuble
aveugle
boeuf
-6me
S6me, s^ment, deuxieme, troisidme, quatrifime, etc.
•fine
-ame
-eine
M6ne, mdnent, egrtee, assene, etc. (and generally all verbal forms in -ener or -6ner)
Hellenes, Helene, etc.
Semaine, fontaine, plaine, chatelaine, douzaine, laine, bedaine, Sedaine, etc.
Peine, haleine, Madeleine, veryeine, baleine, etc.
3. When these seven vowels are unstressed, they are
generally short, the pronunciation gliding easily and to
some extent rapidly over them, as e.g. fhilosophique, muni-
cipalite, politiquement, proiectorat. This is particularly so
in closed syllables, whether the vowel be followed by a
lengthening consonant or not, as corset, expulser, amnistie,
colporteur, parfaitement. Hence we have/or^ [foir], but forcer
[forse] ; dur [dyir], but durcir [dyrsiir] ; Jeuille [foeij], but
Jeuilleton [foejto] ; il se leve [Isiv], but leve-toi [lev twa] ; poivre
[pwaivr], but poivrier [pwavrie] ; car erne [kareim], but
careme-prenant [karem prana]. This rule of course applies to
all the closed syllables in a stress-group apart from the ac-
cented one, inasmuch as a stress-group is one idea, and
thus equivatlent to one word, as la vieille femme [lavjejfam],
tous leurs journaux [tuloer3urno], ce nerf de hceuf [sanerdaboef],
V amour propre [lamurpropr]. Such words as mime, brave,
treize, honnete, naturally preserve their long vowel when at
the end of a stress-group, as le bruit meme [msim], un homme
brave [braiv], nous sommes treize [treiz], une fille honnete
96 MANUAL OF FRENCH PRONUNCIATION
[oneit] ; but this vowel becomes sensibly short if the order
of the words is changed, as le meme bruit [mem], tin brave
soldat [brav], les treize gargons [trez], une honnete fille [onet].
In open pretonic syllables, on the other hand, especially
if initial, these vowels incline to be half-long before the
lengthening consonants [v, z, r]. Thus Paris [pa'ri], raison
[re'zS], saisir [se^ziir], bureau [byro], pourrir [pu'riir] ,irons [i'ro],
souffrira [sufri'ra], armoirie [armwa'rij, plaidoirie (pledwa'ri],
etc. In heros the vowel is fully half-long [cro], while in
heraut it is short [ero]. But even in these cases many
speakers would regard the vowel as ' short,' and the student
must beware against giving it too much length.
In open syllables also (only in a few cases in closed
ones) these vowels are generally half-long in those cases
where they are long in the root-word or stem, i.e. where
they are long when the same syllable is final. In such cases
it is evident that the vowel, being long by nature, may lose
its stress but cannot lose all its length. For example, the
root-words vive and sur both have a long vowel according to
rule. Consequently this vowel preserves some of its length
in the derivatives, as vivant, vivacite, aviver, assure, surement,
etc. Similarly, the vowel is long in the stem-words aime,
naisse, pleur, clair, and so it maintains half-length in deriva-
tives, as aimable [s'mabl], naissant [ue'sq], pleur er [ploe're],
clairement [kls'rma].^ It should be noted, however, that the
half-length disappears if the root-word ends in [3] or [j]
(though these are lengthening consonants), as or age [orai3],
but orageux [ora30] ; courage [kurai3], but courageux [kura30] ;
fouille [fuij], hut fouiller [fuje].
The above cases of half-length in unstressed syllables do
not form a large class comparatively, and many of them
only occur in careful or impressive speech. It should be
understood that in the case of these seven vowels the great
majority of them, when unstressed, are pronounced short.
Indeed, in the case of all vowels, the sound becomes the
^ Examples of this rule are very common. Thus, from the root-
words pierre, serve, terre, breve, we have half-long vowels in pierreux,
empierrer, server, serrure, terreau, terrer, enterrer, brievement, brievete,
etc. We may add adverbs in -erement (as fierement, amerement, etc.),
many verbs in -iser, -ouser, etc., many in -rer (as murer, bourrer, fourrer,
tourer, empirer, etc.), and most verbs in -eurer or -eurrer (as beurrer,
disheurer, icoeurer, pleurer, leurrer, etc.). In verbs in -eurer or -eurrer
the vowel is fairly long, except in demeurer, fleurer, effleurer.
THE VOWELS 97
shorter the farther removed they are from the tonic or
rhythmic Stress. Thus, the a of pate is shorter than that of
pate, and the a of patisserie is shorter still than that of pate.
The same difference is apparent in pdme, pdmer, pdmoison.
Similarly aime is shorter in aime-t-il, where the accent is on il,
than in il aime ; and etes is pronounced more lightly in pous
etes fou than in fou que vous etes ! If there is any doubt as
to the length of any unstressed vowel, let the student pro-
nounce it short, and the chances are that he will be right.
Emphasis may lengthen a vowel in a non-final syllable,
but this is referred to in the chapter on Emphatic Stress.
EXERCISES
1. Pronounce carefully the following words containing half-
long vowels in the pretonic or the root syllable :
Tirer, curer, couvant, briser, trouver, cousant, fleurir, jurer,
bravoure, couru, fetard, brulant, aine, maitresse, clarte, juge-
ment, fraichement, rouler, couler, ecrouler, trainant, maigrir,
aider, baisser, fraisier, tairez.
2. Pick out any half-long vowels in the following list,
giving the reason :
Meunier, viendrai, insecte, embaumer, emmener, cousin,
reposer, labourer, arriver, abreuver, montant, epaissir,
montagne, raisin.
3. Pronounce carefully, with special attention to Quantity,
the following extract, the phonetic spelling of which is given
below for reference :
"Je me souviens qu'etant enfant, je m'etais forme des
idees assez singuli^res du soleil et du ciel. Je croyais que le
soleil se levait derriere une montagne et se couchait dans la
mer, que le ciel etait une voute qui s'abaissait vers rhorizon,
de sorte que je pensais que, si je parvenais j usque-la, je serais
oblige de marcher courbe, sans quoi je me casserais la tete
centre le firmament. J'entrepris un jour d'atteindre a
I'extremite de la voute celeste ; apr^s avoir marche une
heure, voyant qu'elle etait toujours a la meme distance de
moi, j'en conclus qu'il y avait trop loin, mais je n'en restai
pas moins persuade qu'elle existait, et que si je ne parvenais
G
98 MANUAL OF FRENCH PRONUNCIATION
pas a la toucher, c'est que je n'avais pas d'assez bonnes jambes.
Au reste, je me figurais, a la vue des etoiles, que le del etait
perce d'une infinite de petits trous par ou la pluie tombait sur
la terre, comme par un crible, et que les etoiles n'etaient que
la lumi^re de Dieu, qui sortait, la nuit, par ces petits trous.
Cette derniere idee n'etait pas si enfantine." — Bernardin de
Saint-Pierre.
Phonetic Spelling^
39 m suvJE ketaafa, 39 mete forme dez ide ase ssgyljsir dy
solsij e dy sjel. 39 krwajs k9 1 solsij S9 Ive derjeir yn motaji
e s kuje da la msir, k9 1 sjsl etet yn vut ki sabess vEir brizo,
d9 sort9 k9 3 pas£ k9, si 3 parv9n8 3ysk9la, 39 srez obli3e d
marje kurbe, sa kwa 39 m kasre la t£it koitrg I9 firmama.
3atr9pri oe 3uir datEidr a lekstremite d la vut selest ; aprez
avwar marJe yn ceir, vwaja ksl ete tu3uir a la msm distais d9
mwa, 3a kokly kil j ave tro Iwe, me 3 na reste pa mwe persqade
kel egziste, e k9 si 39 n parv9ne paz a la tuje, se k9 3 nave
pa dase bon 3aib. o rest, 39 m figyre, a la vy dez etwal, k9
I9 sjel ete perse dyn efinite d9 pti tru par u la plqi tobe syr
la teir, kom par ce kribl, e k lez etwal nete k la lymjeir d9
dj0, ki sorte, la nqi, par se pti tru. set dernjeir ide nete pa
si ofatin. — bernarde d9 se pjeir.
^ Half-length is not marked in the phonetic spelling, but is regarded
as short.
PART II : THE CONSONANTS
CHAPTER XXII
INTRODUCTORY
105. A Consonant, as already defined (§ 7), is "a sound
caused by friction or stoppage of the breath somewhere in
the mouth," that is to say, it results essentially from some
obstacle, and is thus distinguished from a vowel, which is
a sound modified in the resonance chambers of the mouth,
but having a free unobstructed passage. Thus, [s] and [f]
are examples of consonants produced by friction (owing to
the narrowing of the passage), while [k] and [t] are examples
of stoppage.
106. Consonants are divided into voiced and unvoiced,
and it is essential that the student should clearly understand
the difference. When the expiratory current is accompanied
by vibration of the vocal chords, the consonant is said to be
voiced; otherwise, it is called unvoiced. Thus, [v] is
voiced, the air being set in vibration in passing through the
vocal chords, while [f] is unvoiced, the air being non- vibrating.
Indeed, it may be said that [v] and [f] are voiced and un-
voiced forms of the same consonant. A similar remark
applies to [z] and [s], [b] and [p], [d] and [t], [g] and [k], [3]
and [/]. It is thus evident that consonants go in pairs, each
voiced consonant having an unvoiced one. The consonants
P, m, n, r] are only found voiced as a rule in French, but their
unvoiced forms, denoted phonetically by [1, gi, n, r] occur
occasionally after voiceless consonants owing to assimilation,
as in peuple [poepl], quatre [katr], etc., and are^ found freely
in other languages. Voiced consonants are sometimes called
soft, and unvoiced hard. This is due to the fact that in the
former case part of the force is spent in the vibration of the
vocal chords, and the resulting sound has a soft, musical
quality, while in the latter case the force of the breath is
unweakened and strikes the ear more forcibly.
99
100 MANUAL OF FRENCH PRONUNCIATION
The distinction between voiced and unvoiced consonants
plays an important part in French pronunciation, and cannot
be too strongly marked. Students who have difficulty in
the matter should practise the two forms between vowels.
Thus, [afa] and [ava], [asa] and [aza], [apa] and [aba], etc.
They should then pass on to words, such as the following,
between which there exists no other difference than between
their voiced and unvoiced consonants :
Fer, ver
Foie, voie
Camp, gant
Port, bord
poisson, poison
choix, joie
pelle, belle
coup, gout
les soeurs, les heures
sans savoir, sans avoir
trois cents, trois ans
il tance, il danse
107. In a book of this kind, based on phonetics, the order
of the usual alphabet is meaningless. The consonants must
here be grouped according to the manner of their formation
and the place of their articulation. The following table
represents this grouping, with the unvoiced consonant placed
immediately after its corresponding voiced one. The three
semi-consonants [w, j, j\\ are included, to show their exact
position from the consonantal point of view.
Bi-
labial
Labio-
dental
Dental
1 Pre-
palatal
Palatal \ Velar
Uvular
Glottal
Explosives
b,p
d, t
1 g,k
Nasals
m, IT
n,n
Ji, .P
Fricatives
(w, w)
V, f
z, s
3,1
(3, J) 1 (w, Y)
(q, k) i
h
Trills
r>r
R, R
Laterals
M
(1 mou-
ille)
108. The horizontal divisions, five in number, represent
the five different ways in which the consonants are formed :
(i) Explosives : These are produced by the air passages
being completely closed, and then suddenly opened, resulting
in a kind of explosion.
THE CONSONANTS loi
(2) Nasals : Here the mouth is completely closed, but
the velum is lowered, so that the air escapes through the
nose.
(3) Fricatives : These are due to a narrowing of the
passage at a given point, so that the air passes through
with continuous friction.
{4) Trills : These are produced b}^ the air passage being
closed and then opened several times in rapid succession.
(5) Laterals : Here the passage is closed in the middle
and open at the sides of the tongue.
109. The perpendicular columns show the divisions accord-
ing to the place of articulation, i.e. the place where the
explosion or friction is produced. In the Bi-labial con-
sonants this place is at the lips; in the Labio-dental the
lower lip presses against the upper teeth ; in the Dental
the fore-tongue comes against the upper gums ; in the Pre-
palatal the fore-tongue rises towards the front of the hard
palate, leaving a resonance chamber in the middle of the
tongue (Diagram, § 135) ; in the Palatal the middle of the
tongue (the point being kept down) is bunched up against the
hard palate ; in the Velar the back of the tongue is raised
against the soft palate; in the Uvular the friction comes
between the extreme back of the tongue and the uvula (as
in gargHng) ; in the Glottal the sound is produced by the
air passing through the vocal chords before they are brought
together to produce voice.
The learner should practise the production of the con-
sonants alone, without the adjunction of any vowel, and he
will become accustomed to their proper formation as here
described. Let him, for example, practise [s, t, k], etc., not
as ' ess,' ' tee,' * kay,' but by themselves. French consonants
are all remarkable for their clearness, and any fault in this
respect robs the language of its beauty. To omit or slur over
the [t] in such words as maintenant, atmosphere, or the [p] in
such words as symptome, as even some French people do, is
an inexcusable negligence.
110. We shall now deal with the consonants one by one,
beginning at the bottom of the preceding table and going
upwards. In many cases consonants are not pronounced, and
such cases will be noted. A French speaker does not load
the end of words with consonants. As a rule, therefore, final
102 MANUAL OF FRENCH PRONUNCIATION
consonants are mute, as galop [galo], pr ogres [progrs], lent
[la] ; and if a word ends in three consecutive consonants,
the last two are generally mute, as corps [koir], temps [ta],
rompt [ro], je perds [peir]. Under this rule is included the
verbal ending -nt (in -eitt), as Us ouvrent [uivr], Us allaient
[ale], Us parlent [pari]. In most cases the final consonants
/, /, and in many cases c, r, are pronounced, however, as
tarif [tarif], industriel [edystriel], pare [park], hiver [iveir].
There are, of course, exceptions to these rules ; and as a
general principle it may be laid down that final consonants
tend to be mute in words that are popular and in everyday
use. The more learned or rare a word is, the greater
likelihood is there that its final consonant is pronounced.
CHAPTER XXIII
THE LATERAL CONSONANT [1]
111. For the English [1] the fore-tongue frequently touches
the hard palate too far behind, and in some cases even curls
backwards as in the accompanying diagram. This habit
lessens the clearness of the [1] and of the vowel following.
For the French sound, the tongue should touch the gums
immediately behind the teeth, as in the case of the other
dentals [t, d, n], etc.
Frequent position of tongue Position required for
for English [1] French [1]
Fig. 2
112. Reference has already been made (§ 78) to the ter-
mination -il after a vowel being pronounced with the sound
[j], as corail [koraij], conseil [koseij], fenouil [fanuij], ml
[oeij].i Three other words must be included under this
rule, viz. gentilhomme [3dtijom], gresil [greziij], mW^ [miij].
Similarly, we have referred (§77) to the combination -ill
as being also pronounced [j] after a vowel and [ij] after a
consonant. Thus, grenouille [granuij], fouiller [fuje], que
faille [aij], detailler [detaje], abeille [abeij], veiller [veje],
fille [fiij], griller [grije].
113. In the case of -ill, the sound formerly used was
* I mouille' (represented phonetically by [A]). Some speakers,
^ The termination -oil, however, is pronounced [wal], as in poll
[pwal] .
2 English : millet.
103
104 MANUAL OF FRENCH PRONUNCIATION
who try to use this old sound still, pronounce it [Ij], as
souiller [sulje], piller [pilje] ; and this pronunciation is very
prevalent in some districts of Belgium and very common
throughout France in the four words meilleur, ailleurs, cuillere,
juillet. But this is an incorrect imitation of ' / mouille,' and
confuses such words as souiller and Soulier, piller and pilier,
railler and rallier, fusilier and fusilier, etc. The true sound
of [A], which is a palatal-lateral (see table, § io8), may be
produced by sounding [1] with the tip of the tongue
pressed against the lower teeth. If the tongue be kept
in this position, and the forepart of the blade be raised against
the front ridge of the hard palate, a true sound of [A] is
produced. The consonant is identical with the Italian gli
(Broglie),i\iQ Spanish II (Llerena), and the Provencal Ih [Milhau).
It is still common in Switzerland and Southern France, where
it survives as a provincialism, but it has completely dis-
appeared from normal French in spite of the efforts of orthoepists
to retain it, and the sound used now, as already stated, is the
liquid one [j], heard also for // in Spanish America.
114. There are a few exceptions in which -ill after a con-
sonant is pronounced [il] instead of [ij]. These are chiefly
among the cases where the / in Latin was not followed by
an i, as ville [vil], from villa, tranquille [trakil] from tranquilla.
But wherever there was an i following, the sound [ij] occurs,
as fille [fiij] from filia, famille [famij] from familia, etc. The
[il] sound is the proper one etymologically for the former
group, but [ij] has largely invaded this domain.
Among the cases of the sound [il] may be mentioned the
following : bacille, billevesee, billion, calville, capillaire,
codicille, distiller, instiller, lilliputien, mille, maxillaire,
myrtille, pupille, pusillanime, scille, tranquille, vaudeville,
verticille, ville, with their derivatives. To these must be
added all words commencing with ill- (e.g. illustre, etc.). There
are also several proper names preserving the [il] sound, such
as Achille, Cyrille, Gille, Lille, Mabille, Sillery, etc., and those
beginning with Vill- or Mill- (Villele, Villemain, Millet,
Millerand, etc.). For some words there is hesitation between
[il] and [ij]. Thus, scintiller, vaciller, osciller, titiller, are
heard with both sounds, but [ij] seems to prevail in the two
former [sstije], [vasije], and the [il] sound in the two latter
[osile, titile].
THE CONSONANTS 105
115. We have stated (§111) that [1] is one of the consonants
usually sounded at the end of a word, as in fatal, mortel, del,
col, etc. Note, however, that it is mute in certain cases :
(i) Generally in the termination -il, when not pronounced
liquid, i.e., when coming immediately after a con-
sonant, as haril [bari], chenil [fani], coutil [kuti], fusil [fyzi],
sourcil [sursi]. The following are exceptions, however, in
which the [1] is sounded,^ viz. alguazil, avril, beryl, hissextil, cil,
civil, exil, fil, mil (1000), morfil, piril, pistil, profil, puiril,
sil, subtil, tortil, vil, viril, volatil, le Brisil, le Nil, and two
or three other proper names. Indeed, it may be said that
the [1] tends to be sounded more and more in modern French,
and in addition to the words just mentioned one hears it
frequently now in babil, gril, and even in gresil and mil
(millet) where it is generally [j].
(2) In the words aulne, aulnaie, aulx [0], cul [ky], and com-
pounds {gratte-cul, cul-blanc, etc.), fils, pouls [pu], saoul or
soul [su].2
(3) In all words in -auld, -ault (pronounced [0]) -ould, oult
(pronounced [u]), as Arnauld, Gericault, La Rochefoucauld,
Perrault, Quinault, Herault, Arnould, Guiroult, etc.
(4) Medially in a few proper nouns, as Belfort, Gaultier,
Paulmier, Paulmy, Saulnier, etc.
116. In familiar speech, [1] is often dropped in such words
as il, lis, quelque, celui. In such a case, il retains its [1] before
a vowel, but becomes [i] before a consonant or at the end
of a stress-group, as il y va [iliva], but il part [ipair], s'il vous
plait [sivuple], plait-il [pleti], vient-il [vjeti]. lis becomes
[iz] before a vowel, and [i] before a consonant or at the end
of a stress-group, as its ont [izo], ils vont [iv5], ils sont arrives
[isotarive].^ Quelque becomes [ksk], as quelque chose
[kekjoiz], quelquefois [kskfwa], quelqu'un [kskoe], quelques-
uns [kekzde]. Celui becomes [sqi], as qu'est-ce qu'il me veut,
^ These include words from Latin adjectives in -His.
2 In (2), (3), and (4) the mute / is an ' etymological ' one, introduced
into the language by the old savants.
^ The I beca.me mute in il, ils, as early as the sixteenth century,
except of course in liaison. It was not till the eighteenth century
that sustained speech sought to re-establish it, and it is only education
that maintains it in reading and in careful speech at the present
day.
io6 MANUAL OF FRENCH PRONUNCIATION
celui-ci ? [k8skimv0 sqisi]. The tendency to contract such
words, however, becomes a careless habit, and should be
strictly confined to coUoquial conversation.^
^ Another popular tendency, but much worse, consists in doubling
/ after a pronoun, as : Je Wai vu [39I le vy], tu IV as dit [tyl la di], je
te IVai dit [3tal le di]. This is doubtless after the analogy of il Va vu,
il Va dit, etc., but is not more excusable on that account.
EXERCISES
Louis leur lit la lettre de Lucie.
Lili a lu le livre de Jules.
Voila, la lune et les 6toiles luisent !
La lumi^re de la lampe luit loin.
Le lilas est joli le long de rall6e.
Apr6s l'Ag6silas, H61as ! Mais apr^
I'Attila, Hola ! (Boileau)
Iwi loer li la letra d9 lysi.
lili a ly la li:vr9 da 3yl.
vwala, la lyn e lez etwal
lqi:z !
la lymjsir da la la:p Iqi Iwe.
la lila s 3aU la 15 d lale.
aprs la3ezila:s, ela:s ! mez
apre latila, ola ! (bwalo)
and
[il]
{mixed)
L'anguille 6tait tranquille.
11 entra dans la ville avec sa
bequille.
Elle per9a sa pupille avec unc
aiguille.
J'ai vu le pillage du village.
Sa famille demeurait a Seville.
Les fiUes de Guillaume sont
gentilles.
lagiij et8 trakil.
il atra da la vil avek sa
bekiij.
el persa sa pypil avek yn
eg^V^- .. ^ .,
3e vy la pi]a:3 dy vila:3.
sa famiij damcere a sevil.
le fi:j da gijoim so 3ati:j.
Mute
Fournil, fraisil, gentil, outil,
fenil
furni, frezi, 3ati, uti, fani
11 y a des aulnes a Chatellerault.
Mon fils s'amuse tout son soul.
Le cul-de-jatte m'a donne des
aulx.
Quinault etaitun poete fran9ais.
11 se tate le pouls avant de
partir.
il j a dez om a Jatelro.
mo fis samyiz tu so su.
la kyd3at ma done dez 0.
kino ets oe posit frase.
il sa ta:t la pu ava da
parti:r.
Famil-
iar
11 me semble qu'ils n'ont rien.
11 ne sait pas ce qu'il dit.
11 faut qu'il vienne quelquefois.
Celui-ci peut tuer quelqu'un,
Quelques-uns parlent comme
celui-la.
imsaibl ki no rjs.
insepa skidi.
ifo ki vjsn kskfwa.
sqisi p0 tqe kekoe.
kekzoe pari kom sqila.
CHAPTER XXIV
THE TRILLS [r] and [r]
117. This consonant is pronounced more strongly than in
English, being always ' rolled.' In the South of England and
also in America it is generally ignored altogether, unless it
begins a syllable. For example, farm is pronounced very
much like [faim], fair like [feig], and so on. In the North of
England and in Scotland it is different, the [r] being usually
well * rolled,' but in other English-speaking parts the ' roll '
is never heard except in excited or emphatic speech. The
student who wishes to speak French properly, must there-
fore acquire the French [r], either the palatal one (phonetic
symbol [r]) or the uvular one (phonetic [r]), both of which
are ' rolled.'
118. In the palatal [r], the air is stopped by raising the
point of the tongue to the palate, but the tongue is
continually giving way and reinstating itself in the same
place, so that there is practically a continuous flow of
air, with a kind of trill or trembling. The diagram in
§ 112, for the French [1], illustrates also the tongue position
for [r]. It is the old Roman, as weU as the Italian and the
Spanish sound of the consonant, and being much clearer
than the uvular one, is generally the one used in singing, in
oratorical declamations, and in the French theatres, and is
very acceptable in ordinary conversation. If the learner has
any difficulty in its production, he should practise [tr], as
the tongue position of [t] is somewhat similar to that of [r].
Continuous practice will produce a good palatal [r].
119. The uvular sound [r] is known as the ' Parisian rj
or ' r grasseye.' ^ It is common in Paris and in the large
^ The term grasseyer, however, strictly means 'to speak thick,' to
be unable to pronounce well certain letters, r among others, and con-
sequently it is applied in certain districts to designate unpleasant,
mufl9.ed varieties of [r].
107
io8 MANUAL OF FRENCH PRONUNCIATION
towns of the north and west, and is regarded as more
' aristocratic ' perhaps than the palatal sound, and as belonging
more to cultivated society. The * rolling ' is produced by
the uvula vibrating on the back of the tongue, as in the
accompanying diagram, and the tongue-point, instead of
being raised against the hard palate, is kept down and pressed
against the lower teeth. This r is frequent in the north of
England and in Scotland, giving rise to the Northumbrian
and Scotch burr, and is not difficult to acquire. When one
HAROr/UATB
1SBT7I<
Fig. 3
gargles, for example, a series of uvular r's is produced, and
they are uttered even more energetically, although innocently,
by those who snore. It only remains to sound them without
these accompaniments. Let the learner pronounce [rq], as
if he were going to gargle, and the result will be a Parisian
[r], with the uvula trembling on the tongue. ^
120. One or other of these sounds is absolutely necessary
to good French speech, and a choice should be made between
them. Many excellent teachers insist on the palatal form
and condemn the uvular. The former certainly is the more
widely used, apart from the large towns, and has the advantage
^ The uvular r is largely a modern substitute for the palatal one,
although it is difficult to determine the approximate date when it
became popular. In Mohere's time the palatal form was general in
Parisian speech, as is evident from Le Bourgeois Gentilhomme (II. Sc. 4) :
" Le y se forme en portant le bout de la langue jusqu'au haut du palais,
de sorte qu'etant frolee par fair qui sort avec force, elle lui cede et
revient tou jours au meme endroit, faisant une maniere de tremblement :
Rra." The same description is found in Cordemoy's Discours physique
de la parole (1668). The theory that the uvular r was introduced by
Les Pricieuses has little or no historical support. It is more likely
to have spread as a careless pronunciation of the palatal one, which
requires more energy of tongue, and is not easily learned by children
especially.
THE CONSONANTS 109
of being clearer, more sonorous, and less trying to the throat.
But at all events, the student must produce a proper r,
whether palatal or uvular. If he has a good one already,
he should not change it, as any kind of trilled r is acceptable
to the French.
121. One serious fault with English speakers, as already
stated, is to neglect the r altogether, unless it begins a syllable.
This is particularly noticeable before a consonant in the
interior of a word. Consequently, they pronounce arme like
dme, partir like pdtir, sortez like sautez, etc. Even French
people are found pronouncing parce que as [paskg] ; sometimes
they say bonjou' m'sieur, and they frequently omit the r of
the preposition sur before a consonant, especially I, as sur
le dos [syldo], sur le flanc [sylfla], sur le boulevard [sylbulvair].
This lax pronunciation is the basis of the French conundrum,
" Why is a tanner like the sultan ? " '' Because he is sur le
tan " [sylta]. The fault, which is very common with English-
speaking students, is a serious one, and leads to confusion.
To remedy it, the syllables should be detached from each
other, so as to make the r final, as par-tir. They should
then be pronounced separately, at first slowly and afterwards
more and more rapidly, making sure that the r is energeti-
cally maintained. Practice should be made with such words
as parler, gargon, border, torpeur, cordeau, fermer, nerveux,
mercredi, etc.
122. Another English fault is to insert [9] between a long
vowel and the r, as a glide or bridge between the vowel and
the trill, so that we have pere pronounced as [peiar], tour
as [tui9r], etc. This glide, so common in English, must be
avoided in French, and the vowel followed immediately by the
r, as [peir], [tuir]. (See § 12.)
123. The final r is generally pronounced, as amer [amsir],
cuiller [kqijeir], ether [etsir], Luther [lytsir]. But, since the
beginning of the eighteenth century at least, it has always
been mute in the termination of -er verbs, and the ter-
minations -ier, -yer, -ger,^ -cher of nouns and adjectives
of more than one syllable, as portier [portje], metayer
[metsje], berger [b8r3e], boucher [buje]. To these must be
^ Many foreign proper nouns in -ger, however, have the r sounded, as
Niger, Kruger, CUsinger, Scaliger, etc.
no MANUAL OF FRENCH PRONUNCIATION
added the adverb volontiers [volotje], monsieur [m9sj0],
gars [ga], dejeuner (breakfast), gouter (lunch), diner (dinner),
souper (supper), purler (speech), poulailler, oreiller, and one
or two others.
EXERCISES
Ren6 tarde k lire leur lettre.
Robert regarde les roses rouges,
Rira bien qui rira le dernier.
Richard arrose les arbres fruitiers.
Pierre reviendra mercredi a Paris.
Sa croupe se recourbe en replis tortueux.
(Racine)
Je pars, j'erre en ces rocs ou partout se
herisse, . . . (Lemierre)
rane tard a li:r loer Istr.
robsir ragarda le ro:z ru:3,
ri:ra bje ki ri:ra la dernje.
rija:r aro:z lez arbra frqitje.
pJEir ravjedra merkradi a
pari,
sa krup sa rakurb a rapli
tortqo. (rasin)
3a pa:r, 3e:r a se rok u partu
sa eris. . . . (lamjeir)
[8:r]
-yer
-ger
-Cher
Cher, enfer, fer, ver, fier, hier,
hiver, mer, tiers, acquiers
Jeir, afeir, fe:r, ve:r, fjeir,
je:r, ive:r, m,e:r, tjeir,
akjeir
Aster, tender, revolver, pater
asteir, tadeir, revolvsir,
pateir
Jupiter, Quimper, Esther,
Auber
3ypit£:r, kepsir, Esteir, obsir
Premier, dernier, rosier, singulier,
armurier, barbier, chandelier,
pommier
pramje, dernje, rozje,
segylje, armyrje, barbje,
Jadalje, pomje
Ecuyer, foyer, voyer, noyer,
clayer
ekqije, fwaje, vwaje, nwaje,
kleje
L6ger, menager, Alger, Roger,
Tanger
le3e, menase, al3e, ro3e,
ta3e
Cocher, rocher, toucher, vacher,
Larcher
koje, roje, tuje, vaje, larje
CHAPTER XXV
THE FRICATIVES
124. [f ]. This is an unvoiced consonant, represented generally
by /, but sometimes by ph, especially in Greek words. For
its production, the lower lip is brought into contact with the
upper teeth, but the contact does not intercept the breath
entirely : it only causes sufficient obstruction to produce a
kind of explosiveness when they are separated.
125. Final / is usually sounded. But it is mute in the
following cases :
(i) In clef [kle], and generally in the plurals ceufs [0\} bceufs [b0],
nerfs ^ [neir], and sometimes in cerfs, serfs [seir].
(2) In the combinations : bceuf gras, cerf -volant, cerf-dix-cors,
nerf-de-bceuf, chef-d'ceuvre.
(3) In the place-names : Neuf chateau, Neufchdtel, Neufbourg,
Neuf-Brisach.
EXERCISES
Francois fait une farce au philosophe.
La fille de Frederic est faible.
La foudre frappe la fonderie de fer.
Felix a fusille le fils du forgeron.
La foule effray6e fuit dans le caf6.
Un frais parfum sortant des touffes
d'asphod^les. (Victor Hugo)
Sounded
Chef, bref, bas-relief, canif,
soif, rosbif, oeuf, boeuf,
nerf, serf
fraswa fs yn fars o filozof.
la fi:j da Frederik e feibl.
la fudr frap la fodri da fsir.
feliks a fyzije la fis dy
for3ar5,
la ful efreje fqi da la kafe.
oe fre parfoe sorta de tuf
dasfadel. (viktor ygo)
Jef, bref, baraljef, kanif,
swaf , rosbif, cef , beef, nerf,
serf
1 Many people pronounce ceufs as [0] only after a [z], as in trois ceufs,
douze oeufs, quinze oeujs, but they sound the / in other cases, as in quatre
ceufs, combien d" ceufs, un cent (V ceufs, etc. There is a growing tendency
for final / to establish itself everywhere in pronunciation.
2 In the singular nerf also, the / is mute as a rule when the word is
used figuratively, as in ce style a du nerf [ne:r].
Ill
112 MANUAL OF FRENCH PRONUNCIATION
Ce boeuf gras est de Neuf chatel .
Le Cid est le chef-d'oeuvre de
Comeille.
Les cerfs effrayent vos boeuf s.
Les nerfs sont sensitifs ou
moteurs,
II a mis la clef sous la porte.
Le cerf- volant est un jouet
d'enfant.
Le nerf-de-boeuf est un liga-
ment.
Les oeufs sont tres nourrissants.
sa b0gra £ da noejatel.
l9 sid e la Jsdoeivra da
korneij.
le SEir efreij vo b0.
le neir s5 sasitif u motoeir.
il a mi la kle su la port,
la SErvola st de 3W6 dafa.
la nsrdaboef et de ligama.
lez 0 so trs nurisa.
126. [v]. For this consonant the lip is not pressed so tightly
against the upper teeth, but otherwise the formation is
precisely the same as for /, with the addition of voice from
the vocal chords. In English the amount of voice given is
frequently insufficient, so that there is little difference be-
tween the two consonants, but in French care must be taken
to produce as much voice as possible. Practice should be
made with words beginning with v, the consonant being
lengthened and dwelt upon, so as to ensure full vocal quality.
Thus, v-oulez-v-ous v-enir v-oir v-otre v-oisin ?
Note that w in words of German or Flemish origin is pro-
nounced V, as Waterloo [vaterlo], Weser [vezsir], Wuriemberg
[vyrtebsir]. Wagon is also [vago].
EXERCISES
Voulez-vous venir vivre a Ver-
sailles ?
Voici votre neveu vis-a-vis de vous.
Vivien veut vendre la vieille voiture.
Votre voisin vient voir la ville.
Votre vaisseau arrive vendredi k
Venise.
vulevu vni:r vi:vr a versa:] ?
vwasi votra navo vizavi d vu.
vivje V0 vaidra la vjsij
vwatyir.
votra vwazg vj£ vwair la vil.
votra veso ari:v vadradi ;
vaniiz.
127. [s]. For this consonant, the fore-tongue touches the
upper gums, but a narrow passage is left over the middle of
it, through which the breath is sent with a clear hissing sound.
The articulation ends with an abrupt explosive withdrawal of
the tongue from this position. In French, the s is pronounced
more energetically than in English, the organs being held well
in position during the emission of the breath.
THE CONSONANTS 113
128. Like most consonants s generally remains mute at
the end of words, as bras [bra], toujours [tu3uir], abus [aby],
nous parlons [parlo]. Note that it is also mute in the
interior of some words :
(i) In compounds beginning with the possessive mes-,
or the articles les-, des-, as mesdames [medam], lesquels
[lekel], desquels [dekel].
(2) In names of individuals commencing with Des- or
Bois-, as Descartes [dekart], Despots [depwa], Destouches
[detuj], Boislile [bwalil].
(3) In proper nouns, when it comes before a consonant. In
this case the s plays the role of a circumflex, and causes the
vowel preceding to be long if under stress, as Nesle [neil],
Fresnes [frem], Du Guesclin [gekle], Delisle [dgliil], Chasles
[Jail], Rosny [rony]. In the ordinary spelling, this class
of words has not undergone the change imposed on common
nouns by the Academy in 1740, by which hastir, teste, etc.,
became hdtir, the, etc. The following words, however,
have the s sounded : Montespan, Montesquieu, Pascal,
Islande, Mesmer, Crespin, Robespierre, Esquirol, Isnard,
and two or three others, as well as when s precedes t {Estienne,
Mistral, Restaut, etc.). The s is also sounded in ancient
or foreign names {Israel, Dresde, etc.). In South France,
too, the s has preserved its sound, so that we have Estaing
[ssts], Lesparre [Isspair], etc.
129. Although final s is generally mute, it is pronounced
in the following cases :
(i) In a few words mostly monosyllabic, viz. as [ais], ambesas
dbzais], es [ss], us [ys], ours [urs], hilas [elais], mitis
metiis], jadis [3adis], laps [laps], fils [fis], reps [reps], mars
mars] (the month, or Mars, the god of war),^ sus [sys],
cens [sdis],^ tous (the pronoun, [tuis]),^ parisis [pariziis]
1 When the word is applied to the planet, the s is usually mute, as it
is also in the proper name Cinq-Mars.
2 But in encens the s is generally mute.
3 Note that the s is only sounded in the pronoun, as ils viendront
tons [tu:s], tous debout, etc. This rule prevents confusion between
such expressions as ils ont tous [tu:s] dit and ils ont tout [tu] dit, ils sont
tous plains and ils sont tout pleins, ils oublient tous ce qu'on a dit and
ils oublient tout ce qu'on a dit^ etc.
H
114 MANUAL OF FRENCH PRONUNCIATION
(as in sou parisis), pataques [pataksis]. In this connexion,
note the following variations :
Bis (the adverb) is pronounced [biis], as crier his ; but his
(adjective) is [bi], as du pain his.
Gens is usually pronounced [3a], but one often hears the
pronunciation [301 s] in the south. Gens, a Roman
family, is [381s], as in la gens Fabia.
Lis is now [liis], but the old pronunciation [li] remains
(according to the Academy, at least) in the compound
fieur de lis, and is found in verse.
Moeurs is [moers] now, but the old form [moeir] survives
in verse and solemn speech.
Os is correctly pronounced [os] in the singular, and [o] in
the plural. The pronunciation [os] or [01s] in the plural
is, however, frequently used by medical men in referring
to the bones of the human body.
Plus is [ply] before a consonant, as plus grand, and [plyz]
in Haison before a vowel, as plus aimahle. But it is
pronounced [plys] when final or emphatic, so long as
it is not part of a negation ; thus, metiez deux en plus
[plys], fen ai plus [plys], but je n'en ai plus [ply]. It is
also [plys] in mathematical language, as 2 plus 2 font
4 ; and in the grammatical term le plus-que-parfait.
Sens is generally [sais], as in les cinq sens, contre-sens, non-
sens, en tous sens, il n'a pas de sens, etc. But the s is
sometimes mute in hon sens, always in sens dessus dessous,
sens devant derriere, and generally in le sens commun.
Vis (a screw) is [vis], as is also the compound tournevis ;
but vis, the indicative of vivre and the past definite
of voir, is pronounced [vi].
(2) It is sounded in the foreign terminations (mostly
Greek and Latin) -as, -es, -is, -os, -us, the preceding vowel
being generally long, as atlas [atlais], flores [florsis], gratis
[gratiis], pathos [patois], hlocus [bbkyis] ; also in the Latin
words biceps, triceps, forceps, etc. Chaos, however, is
[kao] ; and the s is not pronounced in the names of certain
cloths : jaconas, lampas, gingas, and damas, nor in balandras,
sassafras, matras, or tetras.
(3) Similarly in proper names, the preceding vowel being
generally long, as Arras [arais], Gil Bias \^\\ blais], LSonidas
[leonidaisj, Midicis [medisiis], Lesseps [lessps], Paris (name
THE CONSONANTS
115
of person) [pariis], Mons [mois], Rheims [reis], La Lys
[liis] (river).^ But when names are prenomens or familiar,
the s remains mute, as in Lucas, Denis, Louis, Alexis,
Nicolas, Thomas, Jisus,^ Judas, Dumas, etc.
130. The letter s is not the only one in ordinary spelling
that represents the sound [s]. The letter c does so before
e, i, y, and g before a, 0, u, as cela, del, cycle, force, glagant,
glagon, regu. The combination sc is pronounced [s] before
e, i, y, as scene [sein], sceau [so], scelerat [sclera], irascible
[irasibl], sceptique [septik]. The letter x is sounded [s] in
a few cases, such as six, dix [sis, dis] when used as nouns,^
soixante [swasait], Auxerre [ossir], Auxois [oswa], Auxonne
[oson], Buxy [bysi], Bruxelles [brysel],* Beatrix [beatris],
Aix-les-Bains [sslebe], Aix-la-Chapelle [sslajapel]. In the last
two, however, the pronunciation [eks] is becoming frequent.
The name of the historical French banker, Law, has gener-
ally been pronounced [lais].^
^ The s is generally mute, however, in those proper names in -ans or
-ens which are pronounced with the nasal [a] {Orleans, Conflans, Le
Mans, etc.), in proper names in -ins {Salins, Vervins, Norvins, etc.), in
Amiens [amjg] and Damiens [damje], in proper names in -rs {Tours,
Flers, Mamers, Vercors, etc.), and in a few others {Soissons, Nyons, etc.).
Metz is [ms:s], and Retz is [re:s].
2 Many Protestants, however, prefer to sound the s of Jisus, so as to
show respect for the name, and perhaps also to make themselves
distinct from the Catholics. Some learned men also sound it in order
to make the name like those of other historical persons.
^ See chapter on numerals.
* The pronunciation [bryksel], sometimes heard, sounds vulgar.
^ This is not due, as some think, to the fact that Laws is a second
form of the name in Scotland, but to the expression ' Law's Bank,'
which was common among the people (cf. chez Maxim's).
EXERCISES
Ces serpents de Perse sifflent souvent.
11 faisait sonner sa sonnette.
Sa sceur Cecile est sans souci.
Ces scies, monsieur, ce sont soixante
sous.
Les cinq soeurs de Simon sont douces.
Voici ces six cent six saucisses-ci.
se serpa da psrs sifla suva.
il faz8 sone sa sonet.
sa soe:r sesil e sa susi.
se si, masj0, S9 so swasait su.
le S8 soeir da simo so dus.
vwasi se si sa si sosis-si.
mes-
etc.
Mesdemoiselles, messieurs,
messeigneurs
Medmwazsl, mesj0,
mesejioeir
ii6 MANUAL OF FRENCH PRONUNCIATION
-s-
mute
Aisne, Quesnel, Asni^res,
Besnard, Cosne, Avesne,
Vosges, Champmesle
Le pain bis est de couleur grise.
La cantatrice a les honneurs du bis.
Sophie s'orne de fleurs de lis.
La France est la royaume des lis.
On 6tudie les moeurs des abeilles.
II lui a donne un os a ronger.
II n'a que les os et la peau.
C'est le plus que je puisse faire.
Un lit, une table, plus deux chaises.
Plus on est de fous, plus on rit.
Je vis a Saint-Servan un escalier a vis.
Ces gens n'ont pas le sens commun.
Cet homme perd le sens du vrai.
sounded
Alo^, cortes, express, vasistas,
stras, iris, oasis, rachitis,
cassis,! albatros, rhinoceros,
burnous, omnibus, pros-
pectus, hiatus, r^bus
Agnes, Bernis, Clovis, Jonas,
Rubens
Ein, ksnel, anjeir, benair,
ko:n, avEin, vo:3, JamEle
l9 pE bi E da kuloe:r gri:z.
la katratris a lez onoeir dy
bi:s.
sofi soma da floeir da li.
la fra:s e la rwajo:m de li:s.
3n etydi le moers dez abEij.
il Iqi a done oen os a r33e.
il na ka lez o e la po.
s£ la plys ka ja pqis fEir.
oe li, yn tabl, plys do Je:z.
plyz on E da fu, plyz 3 ri.
3a vi a s£-SErva oen Eskalje a
vis.
se 3a no pa la sa komde.
SEt am pEir la sa:s dy vr£.
aloEis, kortEis, EksprES,
vasistais, strais, iri:s,
oaziis, rajitiis, kasis,
albatrois, rinoserois,
byrnus, omnibyis,
prospEktyis, jatyis,
reby:s
ajiEis, bErniis, kloviis,
3ona:s, rybgis
! A ' black currant,' but cassis, ' cross drain,' has the final s mute.
131. [z]. This is the voiced consonant corresponding to [s],
as in zele, zoyie, bronze, etc. It requires the same movement
of the organs as [s], but before and during the articula-
tion the vocal chords vibrate, making the consonant soft
and sonorous. In French this vocal quality must be fully
produced.
132. A simple s between two vowels (the first not being a
nasal) is generally pronounced [z], as aise [sze], miserable
[mizerabl], rose [roiz], paisible [pszibl].^ It is not always so,
however, in compound words, whose second element begins
with s. If the first element be any of the prefixes des-, re-
1 English to some extent follows the same rule, as in the words easy,
miser able, rose.
THE CONSONANTS
117
(not re-), mes-, or trans-, the s takes the [z] sound, as desunion
[dezynjo], deshomieur [dezonoeir],^ resister [reziste],^ mesallier
[mezalje], transaction [trazaksjo].^ In the case of pre-, too,
the z sound mostly occurs, as present [prezd], presomption
[prezopsjo], presumer [prezyme].* But in all other cases of
distinct compounds, the s retains its proper sound, as resaluer
[rasalqe], monosyllahe [monosilab], entresol [atrasol], tournesol
[turnesol], antisocial [atisosjal], soubresaut [subraso], Lesage
[l9sai3], havresac [avrasak], vivisection [vivissksjo].
133. Final z is pronounced only in one or two words, such
as gaz [gaiz], fez, and in a few proper names, as Berlioz, Rodez,
Suez, Cortez, etc. In German words it becomes s, as quartz
[kwarts], ruolz [ryols], sedlitz [sedlits].
^ Desuetude is an exception [desnetyd] .
2 Resection and resequer [reseksjo, resske] are exceptions.
3 Transylvanie and transept are exceptions [trasilvani, trasept].
* Preseance, presupposer, and presanctifier are exceptions.
EXERCISES
Voila deux oiseaux sur le gazon.
Mes cousines, Louise et Susanne, sont
jalouses.
Les enfants d'Isabelle sont amusants.
Ces hommes resident dans I'lle de Jersey.
Avez-vous vu les zig-zags des eclairs ?
vwala d0z wazo syr I9 gazo.
me kuzin,. l\vi:z e syzan, so
3alu:z.
lez afa dizabel sot amyza.
sez om rezid da lil da 3erz8.
avevu vy le zigzag dez
eklEir ?
s = [z]
Baiser, maison, nasal, risible,
usurper, Joseph, fusilier,
besoin
beze, m£z5, nazal, rizibl,
yzyrpe, 3oz£f, fyzije,
bazwe
D6s-
Re-
Mes-
Trans-
Pre-
Disaccord, designer, desir,
desoler
Reseau, resider, reserver, re-
soudre
Mesange, mesavenir, mesestimer
Transition, transitif, transiger
Presider, preserver, presenter
dezako:r, dezijie, deziir,
dezole
rezo, rezide, rezerve, rezudr
meza:3,' mezavniir, mezes-
time
trazisjo, trazitif, trazi3e
prezide, prezerve, prezate
s = [s]
Antiseptique, resigner, mansue-
tude, parasol, vraisemblable,
polysyllabe, Lasalle, Lesaint,
Desaix
atiseptik, rasijie, masqetyd,
parasol, vresablabl,
polisilab, lasal, lase, d9S£
ii8 MANUAL OF FRENCH PRONUNCIATION
134. [J] (Ch). This is the unvoiced consonant corresponding
to [3]. In ordinary French spelUng it is represented by the
two letters ch (as in chat, chez, roche) and in many foreign
words by sch (as in scottisch, Herschel, etc), but it is really
a single sound, identical with sh in English {e.g. sheep, shilling,
fashion). It has no relation to the English ch, which is
generally pronounced [tj], as in church, change, etc., and this
sound must be avoided.
135. The proper enunciation of the consonant in French
requires the formation of three small resonators in the mouth,
— the first in the middle of the tongue, the second between
the under part of the tongue and the teeth, and the third
between the teeth and the lips.
Position of organs for [J]. The resonators are marked i, 2, 3
Fig. 4
The chief fault of English students is that they do not
project the lips sufficiently, and they thus suppress an in-
dispensable resonator, the result being that the sound is too
thin and harsh. Practice should be made with chut (silence !),
pronounced [ft], taking care that the lips and other organs
are in position before the breath passes through the mouth.
136. Note that the group ch is not always pronounced [J],
but frequently takes the sound of [k]. This is so in the
following cases :
(i) Always before a consonant, as chrHien [kretje], chronique
[kronik,] Christophe [kristof].
(2) Almost always when final, as varech [varsk], cromlech
[kromlek], Enoch [enok]. In almanach, however, it is
generally mute [almana], and in farouche, punch, Foch, and
Auch (as well as several other proper names of the south)
it is Lf] : [faruj, poij, foj, oj].
THE CONSONANTS
119
(3) Frequently in borrowed words, especially before a, o,
or u, as ar change [arkai3], archa'/que [arkaik], choriste [korist],
icho [eko], chxur [koeir], Antiochus [dtjokyis]. Only in a
few instances does the k sound occur before e, i, as in
orchestre, orchis, lichen, Michel-Ange, Machiavel, etc.
Most words in these three classes are from Greek or Hebrew,
and are easily identified, as there is generally a corre-
sponding English word in which ch has the sound of k (thus
Chretien, Christian, archange, archangel). In all other cases
(these being of the more familiar type) the symbol ch
takes the [/] sound, as chanter, chapeau, chose, archiduc (and
all words compounded with archi-, except archiepiscopal some-
times), archeveque, Michel, etc.
EXERCISES
11 charge les choux sur le cheval.
Charlotte chante dans sa chambre.
Ce chasseur chasse sans chiens.
Les chameaux da marchand sont
choyes.
Charles fait la mouche du coche.
il Jar3 le Ju syr la Javal.
Jarlot Ja:t da sa Ja:br.
S9 Jasoeir Jas sa Jjs.
le Jamo dy marja so Jwaje.
Jarl fs la muj dy koj.
[k]
Technique, Christ, chromatique,
Chloe, chaos, chorus, Chaldee,
catechumene, eucharistie,
psychologic, gutta-percha, loch,
St Roch, Lamech, Munich,
Zurich, Bach, Moloch
tsknik, krist, kromatik, kloe,
kao, koryis, kalde, kate-
kyms:n, 0karisti, psikolo3i,
gytapsrka, lok, S8 rok,
lamek, mynik, zyrik, bak,
molok
[Jl
Chimere, cherubin, fetichisme,
trochee, hierarchic, monarchic
Jiraeir, Jerybe, fetijism, troje,
jerarji, monarji
Christophe cherche du varech.
Saint Roch fut sauve par un chien.
Rachel trouve du lichen sur les rochers.
Charles a cache la table de loch.
L'eucharistie est un sacrament
Chretien.
krist of JsrJ dy varsk.
S8 rok fy sove par oe Jjs.
rajel truiv dy liken syr le roje.
arl a kafe la tabla da lok.
0karisti et oe sakrama kretje.
137. [3] (J). This consonant, which is the, voiced form of
[f], requires the same position of the organs, but with the
addition of vocality. It is indicated in some phonetic systems
by the two letters zh, but it is cne sound only, and is better
represented by the single sign [3]. It occurs in ordinary
120 MANUAL OF FRENCH PRONUNCIATION
French spelling as /, or as g before e, i, y, and is pronounced like
the s in the English words ' leisure," ' measure/ etc., as jour
[3uir], jtiste byst], George [3or3], sage [sai3]. The sound must
not be confused with that of / in English, which is really [d3],
as in John [d3on].
138. Note that in some words the combination ge is pro-
nounced [3], the e mute being simply introduced to soften
the g into [3], since g is always hard before a, 0, u, Thus,
we have geole [301I], pigeon [pi35], gageure [ga3yir], mangeure
[ma3yir], vergeure [v£r3yir], mangeons [ma33]. Similarly, we
have nous jugeons |jy3o], il jugea []y3a], etc.
EXERCISES
Lcs gens sages sont tou jours justes.
Je jouis des longues journees de juin.
Je n'aime pas la neige ni les gelees.
lis ne mangent pas de jambon.
Jules joue avec Jacques tous les
jours.
Ci-git le jeune gendre du geolier.
A dejeuner le juge mangea des
pigeons.
le 3a sa:3 so tu3U!r 3yst.
39 3wi de 15:g 3urne da 3^8.
39 nsim pa la ne:3 ni le 39le.
il n9 ma:3 pa d9 3ab3.
3yl 3u avEk 3a:k tu le 3u:r.
si 31 la 3oen 3a:dr9 dy 3olje.
a de3oene I9 3y:3 ma3a de pi33.
CHAPTER XXVI
THE NASAL CONSONANTS
139. [m], [n]. For the nasal consonants the air is stopped in
the mouth and allowed to pass out through the nose. In m
the stoppage is made by lightly closing the lips, and in n by
raising the point of the tongue to the upper gums. At the
same time the velum is lowered, and the breath thus passes
entirely through the nose. Both m and n are usually voiced
in French, i.e. the vocal chords are set in vibration, and there
is thus a suppressed sound, as it were, which is the intentional
result of an effort to produce voice. Only in exceptional
cases, referred to under Assimilation, do they lose their vocal
quality. The fault of English-speaking students is that they
do not give sufficient voice to these nascd consonants, especially
when they are initial.
140. It should be remembered that final m or n, as
already stated (§ 86 (4)), is sounded in many foreign and
unfamiliar words, and the vowel preceding it has no nasal
sound. Thus, maximum [maksimom], Jerusalem [3eryzal8m],
hymen [imen], dolmen [dolmen] ; also that m is not sounded
medially in automne [oton], nor in damner or its derivatives
{condamner, condamnahle, etc.).
EXERCISES
Mon maitre est mort au mois de mai.
Madame Mimel me semble
miserable.
Monsieur Marcel malmene ma mere.
Maurice aime ce morceau de
musique.
Mon mari est le meilleur du monde.
J 'aime le miel mieux que la moelle.
Le mur murant Paris rend Paris
murmurant.
mo me:tr e mo:r o mwa da me.
madam mimsl ma saibla
mizerabl.
masj0 mars si malmen ma ms:r.
moris e:m S9 morso da myzik.
mo mari £ la msjoeir dy moid.
38:m la mjel mj0 ka la mwal.
la myir myra pari ra pari
myrmyra.
121
122 MANUAL OF FRENCH PRONUNCIATION
Ni Nicolas ni Nanine n'est innocent.
Prenons maintenant notre diner.
Notre niece n'aime pas notre voisin.
Notre journal a des annonces
nombreuses.
Le neveu de Nannette est de noble
mine.
Non, il n'est rien que Nanine
n'honore.
Cette machine d6pense le minimum
de charbon.
Comment fume-t-on I'opium ?
Mon medecin m'a condamn6.
Marie demeure a Menton depuis
I'automne.
II y a un specimen d'un dolmen dans
le museum.
ni nikola ni nanin net inosa.
prano metna notra dine,
notra njes neim pa notra vwaze.
notra 3urnal a dez ano:s n3br0:z.
la nav0 da nanet e da nobla min.
no, il ne rje ka nanin nonoir.
set majin depa:s la minimom da
Jarbo.
koma fymto lopjom?
mo metse ma kodane.
mari damce:r a mato dapqi
loton.
il ja de spesimen doe dolmen da
la myzeom.
141. [ji] (GN). This sound is generally known as 'n mouille.'
It is represented in ordinary French orthography by the
letters gn, as in vigne, agneau, etc., and never occurs initially
in normal speech. As it has no equivalent in English,
beginners are in the habit of turning it into [nj]. Thus, they
pronounce saignait as if it were c'est niais [ssnje], soigne as
if it were soil nie [swanje], and pignon as if it were part of
the word opinion [opinp]. Even many French people have
adopted this pronunciation, as being simpler and easier. But
it is incorrect all the same, and a practised ear at once detects
the fault. For [n] the point of the tongue has to be raised to
the upper gums, whereas for [ji] it must be kept down. More-
over, the sound referred to is a double one [n+j], whereas it
should be a single nasal consonant, although represented
by the two letters gn.
The beginner will have little or no difficulty in acquiring
the true pronunciation, if he will sound the English -ng (as
in sing, ring, etc.) with the point of the tongue kept down
against the lower teeth, and the middle of the tongue raised
and pressed lightly against the hard palate, as in the diagram
on opposite page.
The * bunching up ' of the tongue completely obstructs
the mouth passage, the result being that the velum descends
and the vibrating breath passes through the nose. For the
THE CONSONANTS .. 123
English -ng, it is the back part of the tongue that rises and
comes into contact with the soft palate, but note that for
gn it is the middle that rises and touches the hard palate.
The French sound, while very like the English -ng, thus differs
from it by being articulated farther forward.
Position of tongue for [ji]
Fig. 5
For the correct enunciation of -gn, therefore, all that is
needed is to pronounce it as if it were the English -ng, but see
that the point of the tongue is kept down, and that the middle
(not back) portion is moved upward till it touches the hard
palate. The English word sing, pronounced in this way
(but with close i) will give the French word signe. Similarly,
bang will give bagne ; pang, pagne ; wrong, rogne ; and so on.
Beginners will find the consonant to be most easily pro-
nounced when preceded by the vowel [i], because with [i]
(and still more with [j]) the front of the tongue is raised towards
the hard palate (see § 8), though not sufficiently to close the
mouth passage. If it be raised a little more than for [j], it
will press against the hard palate, thus causing stoppage there
and sending the breath through the nose. A little more
practice with words in -igne {signe, vigne, digne, etc.) will lead
to the correct pronunciation, and words can then be practised
in which the consonant is preceded by other vowels, such as
daigner [dejie], poignard [pwajiair], cogner [kojie], Agnes
[apeis] . When the consonant finishes and the tongue descends,
the sound glides easily into the following vowel. To those
who have no phonetic training, this brief ghde seems some-
thing like a [j], and hence they introduce the semi-consonant,
pronouncing mignon as [mijijo], etc., as they also do with
/ mouille, making it [Ij]. But this is not the right sound, as
the detachment itself of the tongue from the palate makes the
only correct glide.
124 MANUAL OF FRENCH PRONUNCIATION
142. The following table shows the formation of the three
nasal consonants [m, n, ji], together with that of the English
-ng (phonetic sign [g]). In all four, the air is stopped in the
mouth by means of some obstruction and sent through the
nose.
Consonant
Stoppage caused by I Part of palate touched
Closing of lips i None
Point of tongue | Just behind upper teeth
Back of tongue I Soft palate
Middle of tongue i Hard palate, except front part
Note that English has three nasal consonants [m, n, g],
and French has three also [m, n, ji].
143. In a few words, after o or a, the combination -ign-
used to be pronounced [ji], the i being silent. These are words
in which the ancient spelling of -ign for -gn is still retained,
such as poigne [pop], poignee [pope], poignard [popair],
poignant [popa], empoigner [apope], encoignure [akopyir],
oignon [opo], moignon [mop5], Philippe de Champaigne [fapap]
(being the name of the province called Champagne), Montaigne
[motap] (being the common noun une montagne), Cavaignac
[kavapak]. Under the influence of the ordinary spelling,
however, practically every one of these words is almost always
pronounced now the other way, as poigne [pwap], empoigner
[apwajie], Montaigne [motep]. Only oignon seems still to
retain the old sound alone, and would shock good taste if it
were pronounced with [wa].
144. It should be remembered that gn does not always
represent the sound [ji]. When initial (except in gnaf,
gnon, gnangan, and gnognotie, which are words of slang)
and in a few learned words (including their derivatives),
it is pronounced as [g+n]. Thus, gnou [gnu], agnostique
[agnostik], cognition [kognisjo], stagnation [stagnasjo]. This
pronunciation, however, which only introduced itself in the
seventeenth century, shows a tendency to give way in some
cases to [p].
THE CONSONANTS
125
EXERCISES
II accompagne les agneaux aux
montagnes.
Les cygnes du seigneur sont soign6s.
II craignait I'indignation des seigneurs.
L'ignominie est tou jours poignante.
Cette campagne a des montagnes
magnifiques.
Agnes a des oignons d'Espagne.
J'ai vu les vignobles de Joigny.
Les ignorants ne sont pas indignes.
g+n
Agnat, cognat, diagnose, gneiss,
gnome, gnosticisme, igne,
ignicole, ignivore, ignition,
inexpugnable, pugnacite,
magnolier, recognition, stag-
nant, magnificat
il akopaji lez ajio o motaji.
le siji dy SEjice:r so swajie.
il krejie Isdijiasjo de sejioeir.
lijiomini e tu3u:r pojiait.
set kapaji a de motaji majiifik.
Ajie:s a dez ojio despaji.
36 vy le vijiobla da 3wapi.
lez ijiora na so paz ediji.
agna, kogna, djagnoiz, gnes,
gno:m, gnostisism, igne,
ignikol, ignivoir, ignisjo,
inekspygnabl, pygnasite,
magnolje, rekognisjS, stag-
na, magnifikat
CHAPTER XXVII
THE EXPLOSIVES
145. [p], [b]. For [p], the lips are closed (in natural shape)
and then opened abruptly, the compressed breath being
emitted with distinct explosiveness. For [b], the formation
is the same, except that the voiced quality is added.
The fault of English students is that they do not give
sufficient voice to [b], so that one frequently hears something
like [p]. To correct this fault, practice should be made with
the phonetic syllable [oeb], slowly and repeatedly uttered,
taking care that the chords continue to vibrate till the con-
sonant has actually exploded. When this habit of glottal
vibration has been acquired, the vowel [oe] should be sup-
pressed, so that the consonant may be articulated correctly
of its own accord.
146. Note that p is not pronounced in the interior of the
following words 1*. hapteme [bateim], cheptel [Jotel], compte
[koit], prompt [pro], dompter [dote], sculpter [skylte], temps
[ta], exempt [egza], and derivatives from these ^ ; nor in
corps [koir], je romps, tu romps, il rompt [ro], sept, septieme,
septiemement.^ In all other cases it is clearly sounded, as
symptome [ssptoim], adopter [adopte], assomption [asopsjo].*
147. Like most consonants, p and h are usually mute
at the end of words. But the final b is sounded in proper
names, as J oh [30b], nahah [nabab], and in two or three
other words, such as cluh, rob, rumb [roib]. Final p is
1 The consonant p used to be mute in many words in popular use,
especially before t, but only a few cases now remain.
2 The p is pronounced, however, in baptismal, exemption, impromptu,
and sometimes in indomptahle.
3 The p is pronounced in all other derivatives of sept, as septembre
[sEptaibr], etc., which are taken directly from the Latin.
* Sometimes when initial, it falls in very familiar expressions, such as
un [pytit gar f on, etc.
126
THE CONSONANTS
127
sounded in a few monosyllables, mostly foreign, as cap,
Gap (town), cep (generally before a vowel), croup, group,
houp I hop ! and in handicap, hanap, jalap, julep, salep.
EXERCISES
Le bon baron a la barbe blanche.
La robe de la bonne est bleue.
Le boeuf est tombe dans I'abime.
Les bottes de Benjamin sont bnines.
La cabane est batie de briques.
la bo baro a la barb blaij.
la rob da la bon 8 bl0.
la boef E tobe da labiim.
le bot da besame so bryn.
la kaban e bati da brik.
Papa part pour la Peloponnese.
Paul se promene pres du parapet.
Pierre ne prend pas son parapluie.
Le paysan a perdu ses pommes.
Philippe ne comprend pas ce passage.
papa pair pur la peloponE'.z.
pol sa promEn prE dy parapE.
pJEir na pro pa so paraplqi.
la peiza a pErdy se pom.
fiUp na kopra pa sa pasa:3.
11 fait beau temps pour le bapteme.
Le sculpteur est au septieme ciel.
Le dompteur est exempt de blame.
Au bout de compte il rompt ses fers.
11 s'y donne promptement corps et
ame.
il f£ bo ta pur la batEim.
la skyltoeir Et 0 SEtJEm sjeI.
la dotoeir Et egzg da blaim.
0 bu da ko:t il ro se fEir.
il si don protma ko:r e aim.
EUe prend un julep pour le croup.
11 y a un club litteraire a Gap.
Les nababs expedient un group.
11 est arme de pied en cap.
Les ortolans nichent dans les ceps.
e1 pra oe sylEp pur la krup.
il ya de klyb litereir a Gap.
le nabab Ekspedi de grup.
il Et arme da pjet a kap.
lez ortola nij da le SEp.
148. [t]. This consonant is represented in ordinary spelling
b}^ th as well as t. Thus, theme [tsim], athee [ate], luth [lyt],
gothique [gotik], etc. For its enunciation, the air passage is
stopped by the fore-tongue being raised to the upper gums,
and as soon as the stoppage ceases, the compressed breath
escapes with a kind of explosiveness.
149. There is a difference between English [t] and French
[t]. For the English articulation the point of the tongue
touches the hard palate a little behind the teeth (this is par-
ticularly so in the case of [tr]), whereas for the French the
point must touch the gums immediately behind the teeth, as
in the accompanying diagram. The pronunciation of the
English word toe is thus appreciably different from that of
the French word tot, while the English pronunciation of such
a combination as [tr] is so different from the French as to be
128 MANUAL OF FRENCH PRONUNCIATION
particularly displeasing to the French ear. All that the
English speaker needs is to advance the tongue-point a little
more towards the teeth, till it practically comes into contact
with them (compare § 60, and /^ § 112). If he has any
Fig. 6
difficulty, let him imagine that he is going to pronounce a
d, which has the tongue-point nearer the teeth than the
English t, and he will thus come pretty near to the correct
position.
150. Note that the combination -ti (not -ty) followed by
a vowel is usually pronounced [si] or [sj]. It retains the [t]
sound in the following cases :
(i) When initial. Thus, tient (and in compounds, as contient,
retiendra, maintien, etc.), tiede, Hare, etc.
(2) After s or x, as question [ksstjo], combustion [kobystjo],
mixtion [mikstjo].
(3) In those words which have lost s before the t, as chritien
[kretje], old French chrestien ; chdtier [Jatje], old French
chastier; Etiettne [etjsn], old French Estienne ; itiage [etjai3],
old French estiage.
(4) In substantives or adjectives in -tie, -tier, -tiere,
-tieme, as amitii [amitje], pitiS [pitje], portier [portje], litiere
[litjeir], septieme [sEtjsm], Poitiers [pwatje], along with the
adverb volontiers [volotje].
(5) In the verbal terminations -tions, -tiez, and feminine
participles in -tie, if the other forms of the verb have
the [t] sound. Thus, nous sortions [sortjo], vous sortiez
[sortje], nous hdtions [dtp], vous hdtiez [atje], partie [parti].
(6) In a few isolated words : antienne, centiare, corinthien,
etioler, ipizootie, garantie, galimatias, ortier, ortie, partie,
repartie, rotie, sortie, sotie, sympathie, tutie, Claretie, Sarmatie,
Hypatie, Pythie, and all other names with the Greek th.
In all other cases ti is pronounced [si, sj], as suprematie
[sypremasi], nuptial [nypsjal], patient [pasja], initier [inisje].
THE CONSONANTS 129
nous portions des portions [port j 5 de porsjo]. Many of these
cases are spelt alike in English and French, but in English
the ti is pronounced sh.
151. When i is final, it is generally mute, but it is sounded
in some cases, as follows :
(i) In many monosyllables, such as ut, brut, chut [ft], dot,
fat, tut, luth, mat} net, pat, rit, zut, est, ouest, Brest?- The
following special cases should be noted :
But, though normally pronounced [by] in Paris, is some-
times [byt], especially when final or emphasized. Le
hut [by] quHl s'est propose ; but, voild mon hut [byt] I
Fait is generally [fe], but there is a tendency to pronounce
it [fet] when final or emphasized, as fa, c^est un fait
[fet]. It is also pronounced [fet] in the expressions
dire son fait, au fait, si fait, par le fait, vote de fait,
void le fait, il est de fait, je mets en fait, etc. But the t
is never sounded in the plural, nor in tout-a-fait, en fait
de, fait divers.
Soit is pronounced [swat] when used adverbially as ex-
pressing an affirmation or concession ('Be it so '), as
vous le voulez ? Soit. But when it is used in verbal
function, or as a conjunction denoting an alternative,
it is pronounced [swa], as il faut quHl soit juste; soit
Vun, soit V autre.
(2) In learned words, especially those from the Latin, as
accessit, aconit, deficit, exeat, indult, preterit, tacet, transit,
zinith, vivat, etc. Christ is [krist], but JSsus Christ is always
[3ezykri], except among a few Protestants.
(3) In a few foreign words, as ballast, compost, knout, lest.
Loth, malt, raout, toast, volt, whist, yacht [jot, jak], or
[jat],^ entre le zist et le zest, and some foreign proper names,
such as Japhet, Hirat, Rabat, Ghat, Cattigat, Calicut, etc.*
Granit, an Italian word, hesitates between [grani] and [granit].
(4) In the terminations -ct, -pt, as abject, direct, infect, in-
tellect, correct, compact, contact, tact, strict, abrupt, apt, rapt,
concept. Hesitation, however, is shown in exact, suspect,
circonspect, district, in all of which -ct is sometimes mute
1 In mdt (mast), the t is mute.
2 Sept, huit, vingt, cent are referred to in the chapter on numerals.
3 Yacht is a Dutch word, and of the three forms of pronunciation
[yak] is the oldest in French, and perhaps the best.
* Goths is [go], as it is also in its compounds, Wisigoths, Ostrogoths.
I
130 MANUAL OF FRENCH PRONUNCIATION
and sometimes sounded. In aspect, respect, instinct, amid,
the -ct is always mute, as also in les freres Parfaict. The
technical word anspect is [aspek], the t being here an ortho-
graphic error.
152. The student will understand that t is not pronounced
in proper names commencing with the prefix Mont- before
a consonant, as Montpellier [mopslje], Montparnasse, Mont-
rouge, Montpensier. Nor is th pronounced in asthme [asm],
isthme [ism], nor the t in posi-scripium [poskriptom] or
postdater [posdate].
EXERCISES
Ton th6 t'a-t-il ote ta toux ?
Therese tacha de tisser du colon.
Titine, tu tutoies tou jours ta tante.
Tu travailles toute la matin6e.
Tot ou tard, Thomas t'aimera.
La petite Suzette est trop timide.
Tu te repentiras de cet acte.
S'est-il arrete a cote de toi ?
to te tatil ote ta tu ?
tereiz taja da tise dy koto,
titin, ty tytwaij tu3u:r ta
ta:t.
ty travaij tut la matine.
tot u ta:r, toma temra.
la p9tit suzet e tro timid,
ty t9 rapatira da set akt.
SEtil arete a kote da twa ?
ti = [ti,
tj]
Tiers, tierce, tien, un tiens,
digestion, dynastie, amnistie,
bastion, bestial, modestic,
Sebastien, moitie, entier,
frontiere, volontiers, hui-
tieme, quanti^me, p6nul-
tieme, nous gations, nous
Editions, vous invitiez, ex-
ceptiez, batie, rotie, garantie
tJEir, tjers, tje, de tje,
di38stj3, dinasti, amnisti,
bastjo, bestjal, modesti,
sebastje, mwatje, atje,
frStjeir, volotje, qitjem,
katjsm, penyltjem, nu
gatjo, nuz edit] 3, vuz
evitje, ekseptje, bati, rati,
garati
ti = [si,
sj]
Ambitieux, balbutier, ration-
nel, nation, Titien, inertie,
aristocratic, diplomatic, pro-
phetic, essentiel, martial,
differentier, Beotie, Helvetic,
Croatie, spartiate
abisj0, balbysje, rasjonel,
nasjo, tisjg, inersi, aris-
tokrasi, diplomasi, pra-
fesi, esasjel, marsjal,
diferasje, beosi, elvesi,
kroasi, spars j at
Mixed
Nous nous exemptions de re-
tenue au moyen d'exemp-
tions.
Vous n'avez pas I'intcntion
que nous lui intentions un
proces ?
A ces objections nous objec-
tions nous-memes beaucoup
de choses.
nu nu egzatja da rtany o
mwaje d egzapsja.
vu nave pa letasja ka nu
Iqi etatjo de pross ?
a sez ab3eks j a nuz ab38ktj a
numsim boku d Jo:z.
THE CONSONANTS
131
sounded
H6 bien, soit ! La fortune est
mon but.
Zut ! il dira son fait au fat.
Una abstraction n'est qu'un
concept.
II est abrupt mais strict en
affaires,
C'est le chemin direct de Brest.
ebje, swat! lafortyn£m3
byt.
zyt ! il dira so fet o fat.
yn apstraksJD nekde kosept.
il £t abrypt me strikt an
afsir.
s£ I9 Jgme direkt da brest.
153. [d]. For the enunciation of this consonant in French,
the formation of the organs is practically the same as for t,
except that d is voiced, i.e. while the compressed breath is
gathering in the mouth, the vocal chords vibrate, so that
voice is uttered. As in the case of h, this vocal quality is
an important adjunct, and students must see that it is pro-
duced. If they have difficulty, practice should be made
with [oed], until the consonant acquires plenty of voice, when
the vowel can be dispensed with.
154. Final d is pronounced in sud [syd], and in many
foreign words and proper names, particularly those in which
d is immediately preceded by a vowel, as yod, talmud, zend
[zsid], ephod, Alfred, David, Madrid, Le Cid, Bagdad, Porte-
Said, Sind [s8id], etc.^
EXERCISES
Didon dina, dit-on, des os d'un
dindon.
Voici la dcmeure de Madame Didot.
Ad61e devient decidement malade.
Cet endroit est commode pour la
douane.
Daniel a decide de m'en donner deux.
Carlsbad, Conrad, Manfred, Sand,
Leopold, Rothschild, le Sund.
dido dina, dito, dez o doe dedS.
vwasi la damoeir da madam
dido,
a del davje desidema malad.
set adrwa e komod pur la dwan.
dan j el a deside dma done do.
karlsbad, korad, mafred, said,
leopold, rotjild, la soe:d.
155. [k]. For the articulation of this consonant the back
of the tongue is raised against the soft palate, causing com-
plete obstruction of the breathing, and the explosiveness
is produced when this obstruction is relaxed. The part of
1 In mademoiselle, the d easily becomes mute in quick speech, but
the omission of it is liardly correct. As for the pronunciation [mamzel],
it is only used in a familiar or even impertinent sense.
132 MANUAL OF FRENCH PRONUNCIATION
the soft palate touched, or point of articulation, varies
according to the nature of the vowel following. It is farthest
forward for [ki], and recedes gradually for [ke], [ka], [ko],
[ku], but is never so far back in French as in English. In
French, the tongue must keep near the front of the soft
palate. Beginners may have some difficulty in doing this,
but if they will think of [g], which is formed farther forward
in English than [k], they will probably come nearer to the
correct French articulation.
156. As a rule, the letter k occurs in French only in foreign
words (as yak, hock, koran, etc.), but the sound occurs very
frequently, bein^ represented by c (raconter), qu (quatre),
q [coq), X [exces), and ch (chretien) (§ 136). The letter c occurs
most often. Before e, i, y, it takes the * soft ' sound [s], but
it is pronounced ' hard ' [k] before a, 0, u, ou, ce (as car,
cordeau, cure, cou, cceur), immediately before another con-
sonant (as clef, croix, tocsin, accident), and at the end of words
(as bloc, due, pare).
When c comes at the end of a word, it is generally sounded,
as in the words just mentioned (other examples are are, bee,
chic, lac, muse, cognac, cric-crae), but it is mute at the end
of the following words : broc,^ croc, accroc, escroc, raccroc,
estomac, cotignac, eric (jack-screw), lacs, tabae, caoutchouc ;
elere, mare, arc-boutant,^ arc-doubleau,^ Leclerc, Mauclere ;
Saint- Brieue, bec-d'dne (bedan) ; and words ending in -nc
{bane, hlanc, franc, fianc, jonc, trone, vainc, etc.), except zinc,
which is pronounced [zEig], and some proper names such as
Rane [raik]. Note also the following variations :
Done has the c pronounced at the commencement of a sentence,
introducing a conclusion (' therefore '), or when emphatic,
as done [doik] nous nous sommes trompds. But otherwise
it is pronounced [do] (' so,' ' now,' * then '), even before a
vowel or at the end of a sentence, as it est done [do] parti ;
allez done [do] ipargner ees gens-la ; taisez-vous done [do].
^chec has the c pronounced, as suhir un iehee [ejek], des ichees
[ejek] inattendus. But many people do not pronounce the
c when the word is plural, in referring to the game of chess,
as jouer aux ichees [eje], although the suppression of it is
quite out of date.
1 The c is pronounced, however, in de brie et de broc.
2 Architects, however, generally pronounce the c.
THE CONSONANTS 133
Marc (proper name) has the c pronounced when used as a
Christian name, or when referring to the evangelist, as
Saini Marc [sSmark], but the c is often mute in le lion de
Saint-Marc (at Venice), and in the place-name Saint-Marc.
Pore is pronounced [poir] in most cases, but the c is sounded
in porc-ipic [porkepik], and when the word is intended as
an insult, as cet homme est un pore [pork].
157. The combination qu, though generally denoting
simple [k] (as in qui, quel, qualite, quotient, quoique, turquoise,
etc.), sometimes represents [kw] before a or 0 and [kq] before
e or i. This is specially so in learned or foreign words,
most of them from Latin. Thus [kw] occurs in the Latin
quadr- (as quadragesime, quadrature, quadrupede, etc.),^ and
also in many other cases, such as aquatique [a.kwa.tik], aqua-
relle [akwarsl], aquarium [akwarjom], equateur [ekwatoeir],
quaker [kwakr], quartz [kwarts], square [skwair]. On the
other hand, we have [kq] in such words as questure [kqsstyir],
requiem [rekqiem], ubiquite [ybikqite], Quintilien [kqetiljg],
etc.2 In some cases there is hesitation between the
simple sound of k, and one of the other two ([kw] or [kq]).
Thus we have :
loquace ([k] or [kw])
quorum ([k] or [kw])
quasimodo([k] or [kw])
equitation ([k], rarely [kq])
quietisme ([k] or [kq])
6questre ([k] or [kq])
questure ([kq], rarely [k])
quintuple ([kq], rarely [k])
I
158. The letter x is usually pronounced [ks], as in fixer,
vexer, axiome, Alexandre, xylographie, etc. This is so in the
prefix ex- when it comes before a consonant (other than s
or ' soft ' c), as in exclure, extase, explorer, etc. There is a
tendency in careless or popular speech to pronounce the
prefix as [ss] in such cases. Thus we hear esprimer, escuse,
estreme, prendre I'espress, etc. Some teachers may be wrong
in regarding this pronunciation as vulgar, for it certainly
^ The word quadrille, a Spanish word, is [kadri:j].
■ The pronunciation [kw] or [kq] was only introduced into French
about the middle of the sixteenth century in the case of words borrowed
from abroad. Previous to this, the letter u after a guttural was a mere
graphic sign, qu in all cases being sounded as a simple [k] . Consequently,
all words borrowed before this date, and all words of native French
origin, preserve the [k] sound. •
134 MANUAL OF FRENCH PRONUNCIATION
has traditional usage in its favour (the Latin ex- having in
many instances first become es- and then e, as in etrange,
ecluse), but it is a pronunciation that is confusing and strange
to many people, and hardly to be recommended.
When the prefix comes before s or ' soft ' c, it is simply
pronounced [ek], as in exsuder, excellence, exciter, etc. ; and
when it comes before a vowel or h ' mute/ it takes the sound
[egz] or [egz] (see § 31 (3)), as in exiler, exercice, exhausser,
exhorter [egzile, egzersis, egzose, egzorte].^
159. When x is final, it is sounded (being pronounced [ks]) in
nouns when it is preceded by a single vowel in the ordinary
spelling, or a nasal sound, as index [edsks], borax [boraks],
codex [kodeks], larynx [lareiks], sphinx [sfeiks]. But it is
not sounded after a diphthong, as choix [Jwa], choux [fu], paix
[pe], je peux [39P0], Bordeaux [bordo], nor in crucifix, perdrix,
prix, flux, afflux, reflux. It is sounded, however, in Aix
[sks] and Dupleix [dypleks].
^ Also when initial in proper names, x generally takes the [gz] sound,
as in Xavier, Xinophon, etc., although in Ximdnis and X^ris it is
sounded [k], [kimcneis, kereis].
EXERCISES
Claude coupe le crayon avec son
couteau.
Claire 6coute le caquet du coq.
L'ecole technique a quatre classes.
Quand comptez-vous 6crire k
C16ment ?
Combien ces quinze boucles coutent-
elles ?
klo:d kup la krejo avek so kuto.
kleir ekut la kaks dy kok.
1 ekol teknik a katra kla:s.
ka kotevu ekrirr a klema ?
kobjg se ke:z bukla kuttel ?
Mute
Cet escroc a beaucoup de
tabac.
Le clerc a fait des lacs
d'amour.
Le marc de raisin est dans
le broc.
Ces bancs de coraux sont
blancs.
Allez done jouer aux
6checs.
S£t Eskro a boku da taba.
la klsir a fe de la d amuir.
la mair da rezs e da la bro.
se ba da koro so bla.
ale do 3we oz eje.
THE CONSONANTS
135
qu = [kw]
qu = [kq]
ex=[£ks]
ex = [8k]
ex=[egz]
Quai, quadrille, question,
acqulrir, inquiet, li-
quide, claque, Paques
Equation, quatuor, squale,
squameux, quadruple,
quadrant, adequat
Quinquag6naire, ubiquiste,
quietude, Quirinal,
Quinte - Curce, quin -
quennal
Explication, excuse, ex-
patrier, extra, extant,
exposer, expulser, ex-
pansif
Excentrique, excepte, ex-
sangue, exceder, excise,
exces, exciper
Examen, exister, inexo-
rable, exact, exil, exo-
tique, exhumer
ke, kadri:j, kestjo, akeri:r,
ekje, likid, klak, pa:k
ekwasjo, kwatqoir, skwal,
skwam0, kwadrypl, kwadra,
adekwa
kr[ekwa3enE:r, ybikqist,
kqetyd, kqirinal, kqetkyrs,
kyekqenal
eksplikasjo, ekskyiz, ekspatrie,
ekstra, eksta, ekspoze, sks-
pylse, ekspasif
eksatrik, eksepte, eksaig,
sksede, sksiiz, eksE, eksipe
egzams, egziste, inegzorabl,
egzakt, egzil, egzotik,
egzyme
Sounded
Mute
Anthrax, lynx, onyx,
phenix, Styx, Ajax,
Felix, Pollux
atraks, leiks, oniks, feniks,
stiks, a3aks, feliks, polyks
Heureux, faix, taux, je
veux, Bayeux, Meaux,
Morlaix, Trevoux
oer0, fe, to, 39 V0, baj0, mo,
morle, trevu
160. [g]. This consonant is the voiced form corresponding
to the unvoiced [k], and care should be taken that the vocal
chords vibrate fully during its emission.
It should be remembered that the phonetic symbol [g]
does not represent the ' soft ' sound of g (before e, i, y), this
being denoted by [3], but represents only the ' hard * sound,
which occurs before a, 0, u (as galop, fagot, goulu, aigu),
before e or tin foreign names {Hegel, Gibbon, etc.), immediately
before another consonant except n (as grand, suggerer, Bagdad),
and at the end of words (as grog, whig, Zadig).
161. The combination gu followed by i or e is not always
pronounced [gq]. As a rule, it is simply [g], the letter u
136 MANUAL OF FRENCH PRONUNCIATION
being interpolated to harden the sound, as gue [ge], guise
[giiz], anguille [Qgiij], tongue [loig], begue [beg]. But it is
pronounced [gw] before a in a few learned or foreign words,
as guano [gwano], la Guadeloupe [gwadlup], Ungual [legwal] ;
and [gq] in the verb arguer, and before i in all derivatives
from the stem aigu-, as well as before i in a few learned
or foreign words, as arguer [argqe], aiguille [egqiij], amhiguite
[abigqite], linguiste [legqist], Guyane [gqijan], Guy on [gqijo].
But in aiguiser there is hesitation between [gq] and [g], as
there is also in Guise (proper noun).
162. The letter g is not pronounced in the interior of
some words, such as sangsue [sasy], signet [sine], vingt [ve],
vingtieme [vet j em], doigt [dwa], doigter [dwate], Longwy
[lowi]. Legs (a legacy) is generally [le], but many French
speakers now pronounce the g [leg], which however is a mere
adventitious letter due to a false etymology (the word is
from laisser, not from the Latin tegatum). But as the word
is a technical and juridical one, perhaps the pronunciation
[leg], which respects the orthography, is the better of the
two.
163. At the end of a word g only occurs as a rule after
a nasal vowel, or in bourg and its compounds. It is
generally mute, as long [13], rang [ra], hareng [ard], bourg
[buir], faubourg [fobuir], Cherbourg [ferbuir].! But it is
pronounced in a few foreign words, as grog, whig, zig-zag,
gong [goig], Liebig, lasting [last§ig], pouding [pudeig]. The
word joug is correctly pronounced [3ug], although the pro-
nunciation [3u] and [3uk] are also widespread.
164f. Note that c is pronounced [g] in second and derivatives
{seconder, secondaire, etc.), in zinc, and generally in the com-
pound word reine-claude [ren-gloid], although in this last
case the c is more and more asserting itself under the influence
of orthography.
^ But bourgmestre is [burgmsstr], and Bourg (name of place) is [burk].
THE CONSONANTS
137
EXERCISES
Le gars Gaspard s'est 6gar6.
C'est une guerre longue et grave.
La gondole est guid6e par Guillaume.
Gustave a 16gu6 ses gains aux gueux.
Regardez le gros tigre k la grille.
la ga gaspa:r set egare.
set yn ge:r 15:g e graiv.
la gSdol e gide par gijoim.
gystaiv a lege se ^g 0 g0.
ragarde la gro tigr a la
griij.
gu = [g]
Gu^re, distingu6, orgue, dogue,
langue, narguer, briguer
ge:r, distlge, org, dog,
la:g, narge, brige
gu = [gw]
Alguazil, Guatemala, Guadal-
quivir
algwazil, gwatemala, gwa-
dalkiviir
gu = [gil]
Aiguilleur, contiguity, exi-
guity, arguant, inguinal, lin-
guistique
egqijoeir, kotigqite, egzi-
gqite, argqa, egqinal,
legqistik
Mute
J'ai achet6 des harengs dans le
bourg.
Mettez le signet k la vingti^me
page.
11 avait des sangsues sur le
doigt.
11 tient le premier rang k
Cherbourg.
Le march6 de ce faubourg est
long.
3e ajte de ara da la bu:r.
mete la sine a la vet j em
pa:3.
il ave de sQsy syr la dwa.
il tjg la pramje ra a
Jerbu:r.
la marje da sa fobu:r e 15.
Sounded
L'ivrogne marche en zig-zag.
Les whigs sont partisans de la
liberte.
Le son du gong retentit au loin.
Je ne les aime pas : ils boivent
du grog.
Beaucoup de legs deviennent
des jougs.
livroji marj a zigzag.
le wig s3 partiza da la
liberte.
la s3 dy g3:g ratati 0 lw§.
3a n lez e:m pa, il bwa:v
dy grog,
boku da leg davjen de 3ug.
CHAPTER XXVIII
THE LETTER H
165. This letter, when sounded, is a glottal fricative,
i.e. it is the sound of the air passing out through the
glottis before the vocal chords begin to vibrate for the
following vowel. But it is not pronounced in France,
except in a few provinces, such as Normandy, Lorraine, and
Gascony. In these parts la halle, une haute montagne are
pronounced [lahal], [yna hoit motaji]. Sometimes also
Parisians unconsciously sound the h to avoid a hiatus, as
la-haut [laho], and even insert it for the same purpose where
it does not occur in writing, as fleau [fleho], cent un [sahde].
Normal speakers sound it in the interjections aha ! oho !
etc., and usually in certain words which require to be uttered
energetically, as je hais, c'est une honte, it est tout haletant
(compare * hattention ! ')
But apart from the above instances, h is never pro-
nounced now in good French. It has disappeared from
cultured speaking since the middle of the seventeenth century,
and only the sign h survives in the spelling.
166. In spite of this fact, there are two varieties of h.
i These are generally called * h |nutgj and * h aspirate,' but
% , jrv these names are misleading, as they imply that h is sometimes
yj J^ sounded. Better terms would be ' h conjunctive ' and
• O,"^ * h disjunctive,' for the sole difference between the two is
• ^ that the former is treated as if it were jion-existent and thus
f ^^f does not prevent elision and liaison, whUe the laffef is treated
.^* as a regular ^consonaint anrl >4^eyents these things. For
i/- ^ instance, m neither of the words homme and heros is the h
sounded, but the former is ' mute ' or conjunctive, and hence
-V , we say V homme, les hommes [lez om], d l' homme, while the
vy-Y A^ latter is ' aspirate ' or disjunctive, and so we say le heros [b
. r ero], les heros [le ero], au heros. Similarly, we have un habit
THE CONSONANTS
139
[oen abi], but un hameau [oe amo] ; en Hispanic [an ispani],
but en Hollande [a olaid] ; en eau [an o], but en haul [a o].
The same remark applies to h in the interior of words. Thus,
we have enherher [dnerbe], with the h 'mute,' but enhardir
[aardiir], ahuri [ayri], dehors [daoir], with the h 'aspirate.'
167. To the ordinary student, there is little means of
knowing when an /^ is * aspirate.' There are nearly four
hundred words of the kind in French, which allow neither
elision nor liaison before them. As a rule, if a word
beginning with h be from the Latin or Greek (this
includes all in hy-, as hyperbole, etc.), the h is * mute,' but
otherwise it is ' aspirate.' Hence we have Vhomme, I'herbe,
Vhabilete, which are from Latin roots, but la harpe, la honte, la
haine, which are of Teutonic origin. The principal exception
to this rule is heros, in which the h is * aspirate,' but all its
derivatives have h ' mute.' Thus, le heros, du heros, un heros
[ce ero], but V heroine, de Vhero'isme, un hero'ique exemple [den
eroik egzaipl] . The following list of words with h * aspirate '
may be useful. To this must be added all derivatives from
these, and most foreign names beginning with h.
ha!
hanap
haridelle
heler
habler
hanche
harnais
hem !
hache
handicap
hare
hennir
hagard
hangar
harpe
Henri
haie
hanneton
harpie
Henriade
haillon
hanse
harpon
heraut
haine
hanter
hart
h^re
hair
happer
hasard
herisser
haire
haquenee
hase
hernie
hale
haquet
hate
heron
haleter
harangue
haubans
heros (not de-
haler
haras
haubert
rivatives)
halle
harasser
hausse
herse
hallebarde
harceler
haut
hetre
hallier
hardes
have
heurt
halo
hardi
havre
hibou
halte
harem
Havre (Le)
hie
halt^re
hareng
Haye(La)
hideur
hamac
Harfleur
hel
hierarchic
hameau
hargneux
heaume
hisser
hampe
haricot
hein?
hoi
140 MANUAL OF FRENCH PRONUNCIATION
hobereau
horde
houri
huguenot
hocher
horion
hourra
huit
hochet
hors
houspiller
hulotte
hol^l
hotte
housse
humer
homard
houblon
houx
hune
hongre
houe
hoyau
huppe
honnir
houille
hublot
hure
honte
houle
huche
hurler
hoquet
houlette
hue !
hussard
hoqueton
houppe
huer
hutte
EXERCISES
H
' Aspirate '
Harold coupe la haie avec la hache.
La horde s'enfuit k la hate.
ie haranguais les dames de la halle.
,es hiboux hantent la hutte.
Je hais ce hussard : il hurle trop
haut.
Le havresac de ce h6ros est en haillons.
Le hibleur est hardi et honteux.
II y a des houblons pr^ du hameau.
arold kup la 8 avek la aj.
la ord s afqi a la a:t.
39 arSge le dam da la al.
le ibu art la yt.
39 E S9 ysa:r, il yrl tro o.
la avrasak da sa ero et a aj 3.
la abloe:r e ardi e 3t0.
il j a de ubl3 pre dy amo.
CHAPTER XXIX
THE NUMERALS
168. The six numerals cinq, six, sept, huit, neuf, dix, require
special reference, so far as their final consonants are concerned,
inasmuch as these consonants are sometimes mute and some-
times sounded. When these numerals are used as nouns,
or in dates,^ the final consonants are always sounded (x
in this case being pronounced [s]), as [sEik, sis, set, qit, noef,
dis] (see first and second columns below). When, however,
they are used as adjectives, their final consonants are mute
before another consonant or h ' aspirate,' but are carried
forward in liaison before a vowel or h * mute,' x in this case
becoming [z], and / becoming [v]. (See third and fourth
columns.) Thus :
Nouns 2
In dates
A djectives
{before con-
sonant)
A djectives
{before vowel)
11 en^a cinq [sesk]
lis sont six [sis]
Combien ? Sept
[set]
J 'en ai huit [qit]
6tez sept [set] de
neuf [noef]
Ajoutez-en dix
(dis)
Le cinq [se:k]
mars
Le six [sis] avril
Le sept [set] mai
Le huit [qit]
juin
Le neuf [noef]
juillet
Le dix [dis]
aout
Cinq [se] francs
Six [si] kilos
Sept [se] chaises
Huit [qi] livres
Neuf [noe] ^ mai-
sons
Dix (di) ho-
mards
Cinq [sEik] amis
Six [siz] hommes
Sept [set] oeufs
Huit [qit] oies
Neuf [noev] ^
ans
Dix [diz]
heures
1 When used in dates, the numerals are really nouns. Le neuf mai,
e.g., is in reality an abridgement for le neuf de mai.
* Similarly, le six [sis] de cceur, le sept [set] du mois, Charles VIII
[qit], le neuf [noef] de cceur, un dix [dis] en chiffres, etc., where the
numerals are nouns.
^ Neuf before a consonant, and also when the / becomes [v] in liaison,
is pronounced with close vowel [0] by many people, as neuf maisons
[n0 mezo], neuf ans [nov a].
141 .
142 MANUAL OF FRENCH PRONUNCIATION
The word neuf, however, when used as an adjective
before a vowel, does not always change / into [v]. It does
so in neuf ans, neuf heures, and frequently neuf hommes,
but there is a tendency in all other combinations to preserve
it as [noef], as in neuf amis, neuf enfants, etc. The word
sept shows the same tendency when used as an adjective
before a consonant. In this case, instead of [se], many good
speakers pronounce it [set], so as to avoid confusion with
ces, ses, as sept chaises [set Jeiz].
169. The numeral vingt follows the same rules as the above
six, except that when used as a noun its final consonants
are generally mute, so that we have il y en a vingt [ve], le
vingt [veit] mai, vingt [ve] femmes, vingt [veit] abricots. Note
also that, contrary to rule, the t is sounded in the numbers
twenty-one to twenty-nine, although not in compounds, as
vingt-deux [vetd0], vingt-cinq [vetseik], but quatre-vingt-deux
[ved0], quatre-vingt-cinq [veseik].^
170. The numerals deux, trois, cent have their final con-
sonants mute in all conditions except when coming as
adjectives before a vowel or h ' mute.' In the latter case liaison
takes place, and the final consonant is carried forward, as
deux heures [d0z oeir], trois amis [trwaz ami], cent hommes
[sat om], deux cents etoiles [6.0 saz etwal]. It should be
remembered, however, that while liaison and elision thus
take place after numerals, none can take place before them,
so that we must say les deux onze [d0 5iz], les trois huits
[trwa qit], cent un [sa de], cent unieme [sa ynjem], etc.^
As the remaining numerals, except un, end in e mute,
their final consonants occasion no difficulty.^
^ The reason for this is that if the t were sounded, e.g. in quatre-
vingt-deux, this would mean four times twenty-two, instead of four
times twenty, plus two.
2 But liaison takes place in dix-huit and in the combination mesure
d, six-huit [sizqit].
3 In olden times, when the final consonant was pronounced in all
nouns of number (including vingt, deux, trois, etc.), un was pronounced
[oen]. Even yet musicians beat time by une, deux [yn, d0], time is
marked at drill by une, deux [yn, d0:s], and we have such expressions
as ne faire ni une ni deux, en donner une ^ quelqu'un, etc., which are
survivals of the old pronunciation.
The numeral un should be distinguished from the article un. The
former, as indicated above, allows no elision or liaison before it, as
le un et puis le cinq [l9 oe], quatre-vingt-un [katraveoe].
THE NUMERALS
143
EXERCISES
Le cinq mai j'ai gagne six francs.
Voulez-vous neuf livres ? Je n'en ai
que six.
Le huit mars nous 6tions neuf.
Ces cinq epingles valent dix centimes.
Le six aout dix ou douze sont arrives.
Otez cinq de huit, reste trois.
J'ai cinq cigares et dix allumettes.
Le neuf mai, elle aura neuf ans.
Voici sept jouets pour les sept en-
fants.
II est mort le dix avril, k six heures.
J'ai huit poires et huit abricots.
Le sept Janvier nous en avons vendu
sept.
Voici neuf ceufs pour les neuf hommes
I9 sE:k me 36 gajie si fra.
vulevu nee li:vr ? 39 nan
e k sis.
la qit mars nuz etjo noef.
se sek epeigla val di satim.
\d sis u dis u du:z sot arive.
ote s£:k da qit, rssta trwa.
3e se siga:r e diz alymet.
la ncEf me, el ora ncev a.
vwasi se 3we pur le set afa.
il e mo:r la dis avril, a siz oeir.
3e qi pwair e qit abriko,
la set 3avje nuz an avo vady
set.
vwasi noef 0 pur le ncev om.
Exercises on vingt
Vingt soldats ont ete blesses.
Maurice m'a donne vingt ecus.
Vous avez quatre-vingt francs, don-
nez-m'en vingt.
J'ai compte vingt-huit oiseaux.
ve solda 5t ete blese.
moris ma done vet eky.
vuzave katrave fra, donema
ve.
3e kote vetqit wazo.
Exercises on deux, trois, cent
Les trois hussards ont pique des deux.
II y a cent soldats dans la caserne.
Mes deux amis partent h. trois heures.
Les trois hommes ont plante cent
arbres.
II a prete k cinq pour cent.
le trwa ysa:r 5 pike de do.
il j a sa solda da la kazern.
me doz ami part a trwaz oe:r.
le trwaz om 5 plate sat arbr.
il a prete a se pur sa.
No Liaison
II y avait quatre cent onze hommes.
II a appris la page cent un.
J'ai quatre-vingt-un francs.
il j ave katra sa 5:z om.
il a apri la pa:3 sa de.
3e katraveoe fra.
CHAPTER XXX
DOUBLE CONSONANTS
171. As a rule, the double consonants of common spelling
are simply pronounced as a single one.^ Thus, grammaire
[grameir], ahhe [abe], affaire [afeir], nommer [nome]. Other
examples are :
Accroc, allee, arret, annee, abbaye, accabler, addition,
aggraver, appas, commerce, commis, casser, fourrure, prudem-
ment, sommet, souffrir.
The doubling of [j] in such forms as nous croyions [krwaj-jo],
vous voyiez [vwaj-je] is held by some teachers to be artificial
and pedantic. But the best speakers undoubtedly double
it, even in rapid speech, pronouncing croyions [krwaj-jo]
as distinct from croyons [krwajo].
172. At the same time, ' double ' consonants occur
frequently in French, although not so often as in Norwegian
or Italian. They are not really double, however, in the sense
of being uttered twice, but are merely a lengthening or
prolongation of the sound. That is, the commencement of
the consonant, instead of being immediately followed by
the explosive finish, is separated from it by an interval
more or less long, during which the air presses with increasing
force against the obstruction. Compare the English words
wholly, meanness, oneness, missend, etc. Such double con-
sonants in French fall under four categories :
(i) Certain grammatical forms, as je courrai [kurie],^
nous mourrons [murio], je requerrai [raksrie]. Under this
class are included the futures and conditionals of courir
1 The same rule applies in English, as ejfort, abbey, litter, etc.
* It is not usual in phonetic transcript to indicate lengthening of
consonants, but the sign (:) may be used for this purpose, or the con-
sonantal symbol doubled.
144
THE CONSONANTS
145
(and derivatives), of mourir, and of compounds of querir
(acquerir, requerir, etc.). The lengthening in such cases is
a means of distinguishing tenses, as il mourait and il mourrait,
nous courons and nous courrons.
(2) Words in which e mute is elided, thus bringing
the two consonants together, as nettete [netie], extremement
[ekstremid], il server a [sen a]. Similarly, one sometimes hears
tout a Vheure popularly contracted to [tia loeir]. If there is
any difficulty in * doubling ' the consonant in these cases,
one is better by far to sound the elided vowel than to produce
such combinations as nete, extrement, etc., or to say la dent
for Id-dedans.
(3) When two words meet, the first word ending and
the second beginning- with the same consonant, as robe
blanche [rob bldij], chaque quartier [fak kartje], partir resolument
[partiir rezolymd], pour rien [pur rje]. It is very important
that such consonants should be well lengthened, otherwise
confusion results. In the following table, for example, there
is nothing to distinguish one expression from the other except
the lengthening of the consonant :
Pas 9a
lis m'ont tous
Les jetees
Les doux zouaves
Chaque abri
(4) In the interior of learned or rare words, par-
ticularly in the case of the letters /, m, n, r, s, t, p, d, as malleable
[malieabl], mammifere [mamiifeir], decennal [dessnial], errer
[erie], iessiture [tesiityir], attique [atiik], appetence [apietais],
reddition [rsdiisjo]. In this class are included, as a rule,
many words beginning in ill-, imm-, inn-, irr- (all four
representing the Latin prefix in-), and a few in com-, as
illegal, illegitime, immigration, immortel, innervation, irreligion,
irruption, commuer, etc. Further examples are here given :
[1 :] Allah, Apollon, allegorie, allusion, appellatif, alliteration,
belliqueux, belligerant, collateral, collision, constellation,
ellipse, foUicule, gallicisme, parallele, pellicula, maxillaire,
syllabe, solliciter, villa.
[m :] Ammon, Emma, Emmaiis, gamma, mammaire, mammoth,
sommaire, summum.
[n:] Annuel, annexe, Anna, annihiler, annotation, annular,
11 1'a dit
11 a dit
Passe 9a
Toute triste
Tout triste
lis montent tous
Une noix
Une oie
L'ai-je jete
Elle lut
Elle eut
Les douze zouaves
Celle-la
C'est la
Chaque cabri
^
146 MANUAL OF FRENCH PRONUNCIATION
connexe, connivence, cannibale, Cinna, innocuite, innom^,
Linn^, Porsenna.
[r :] Irrespectueux, horreur, terreur, narrer, errata, corregidor,
Pyrrhus, Urraque.
[s :] L'Assyrie (distinguished from La Syrie), assoupir, assouplir,
asservir, assomption, classique, dissemblable, Nessus.^
[t:] Attitude, atticisme, battologie, dilettante, guttural, ven-
detta.
[p :] [d :] Appius, adduction, Adda, Eddas, quidditd, appendice,
appetence, appogiature.
173. In the first three categories above, the doubling
or lengthening of the consonant is obligatory, and should be
quite marked. In the last category, however, there is
hesitation. Indeed, many of the words there mentioned
are pronounced with a single consonant by a large number of
speakers. The doubling of a consonant in the interior of a
word, unless to avoid confusion, is contrary to the tradition
of the French language, which has simplified all double
consonants in words of native origin. The lengthening
should be strictly confined to learned, rare, or special
words ; it is distasteful and pedantic when employed in words
which are in eveiyday use and have become familiar. Hence
many good speakers pronounce the consonants as single,
not only in many of the cases above, but especially in such
words as :
Immense, immeuble, immoler, illustre, innovation, illogique,
inne, impeccable, alleguer, annales, college, commentaire,
commiseration, effraction, hippodrome, etc.
If any rule of guidance is needed, it may be said that the
doubling of the consonant should be employed in particular
when one wishes to express emphasis, or some emotion
such as disgust, fear, admiration, irony, anger, etc. Thus, we
may express emphasis in lit-terature ; disgust in il me fait
hor-reur ; fear in 0 ter-reur 1 ; admiration in cet il-lusire
savant ; elegance in as-soupli ; etc. But otherwise the
doubling of the consonants under the last category is not
frequent in good French, except among rather illiterate people
1 Cases where s is pronounced double should not be confused with
ordinary double s in such words as mission, richissime, massacre, where
the -ss- is merely the orthographic sign of unvoiced s.
THE CONSONANTS
147
who affect to speak well, and upon whom the ordinary
orthography has a growing influence in this matter as in
many others (such as final consonants, etc.).
EXERCISES
I
Je mourrai, tu mourras, il mourra.
Le cheval ne courra pas aujourd'hui.
II encourrait mon indignation, s'il
faisait cela.
II s'enquerra de la v6rit6 du fait.
Nous nous entre-secourrons toujours.
Le g6n6ral reconquerra cette pro-
vince.
Nous recourrons k la cl6mence du roi,
II mourrait, si vous ne le soigniez pas,
Je vous requerrai d' insurer toute ma
r6ponse.
II acquerrait des talents, s'il etait
diligent.
II vous requerra de partir, si vous
I'insultez.
de
II
cette
Vous tirerez satisfaction
injure.
II barrerait cette route, si vous
I'ofEensiez.
La verrerie est I'art de faire le verre.
L'embaumement 6tait pratiqu6 par
les figyptiens.
II serrerait la recolte, s'il faisait beau
temps.
Ill
Cet Arabe basan6 a une haute taille.
II faisait du soleil hier.
Voil^ des mceurs singuli^res.
J'aime mieux une lame mince,
Donne-nous I'histoire romaine,
Tu f aches Charles, ote-toi d'ici.
Remarque que tu te trompes.
II passe son temps k lire.
Jeanne ne coupe pas dedans.
39 murre, ty muria, il mur:a.
la Javal no kur:a pa o3urdqi,
il Qkurie mon gdijiasjo, sil
faze sala.
il sakeria da la verite dy fet.
nu nuz atras9kur:3 tu3u:r,
la 3eneral rakokeria sst proveis.
nu rakur:3 a la klemais dy rwa.
il mur:s, si vu na la swajije pa.
3a vu rakerie d esere tut ma
repois.
il akerie de tala, sil ets diH3a,
il vu raker :a da parti :r, si vu
lesylte.
vu tirje satisfaksjS da set
g3y:r.
il bane set rut, si vu lofasje.
la ver:i e la:r da fe:r la ve:r.
labom:a ete pratike par lez
e3iptje.
il ser:e la rekolt, sil faze bo ta.
set arab bazane a yna oit ta:j,
il faze dy soleij jeir.
vwala de mcers segyljeir.
3e:m mj0 yn lam me:s.
doniu listwair romen,
ty fa: J Jarl, o:t:wa disi.
ramark ka ty ta tro:p.
il pa:s son tQ a liir.
3a:n na kup pa dada.
y
PART III
WORDS IN COMBINATION
CHAPTER XXXI
ELISION
174. So far we have considered the pronunciation of
separate words, independent the one of the other. But this
is not enough for correct speech, as many words, when they
take their place in a sentence, require their pronunciation to
be modified according to their position and relative importance.
Individual words have to form themselves into groups, and
groups into sentences, all which leads to changes in the sounds.
Thus, the word semaine by itself is pronounced [samsn] in
the dictionary, but if preceded by the article it becomes [smsn]
in conversation ; de by itself is pronounced [d9], but it
becomes a simple [t] in chemin de fer, de temps en temps, etc. ;
and magnifique, like all French words, has normally the
stress on the last syllable, but when emphasized, as in vol
magnifique, it takes the stress on the first. We may say
generally that almost every sound is influenced to some extent
by neighbouring sounds. There is not merely juxtaposition,
but to some extent ' compenetration ' ; and hence it is not
sufficient to know the pronunciation of individual words :
their pronunciation in relation to each other has to be
understood. In the following chapters we shall explain what
modifications of sound become necessary when words are
formed into connected speech.
175. The most numerous and important changes which
take place in the pronunciation of words are due to changes in
their form, owing to the elision of the so-called e mute (or
e caduc) sound. In many cases this indeterminate sound
is quite clearly pronounced, as in le jardin, demain, de pres,
etc. ; but at other times it happens to come into certain
positions where it is usually elided, at least in the general
149
150 MANUAL OF FRENCH PRONUNCIATION
flow of conversation, as dans le jar din [dd 1 3ardE], apres-
demain [apre dme], tout de prds [tu d pre]. In some cases
the elision is apparent in the ordinary speUing, the letter
itself being omitted (e.g. Vhomme, qu'avez-vous, etc.), but in a
much larger number of cases the letter remains while the sound
is elided. The fact is that the e mute, which only occurs in
open unstressed syllables, is of such little consequence that it
is suppressed as often as convenient. The question of when
to suppress it is one of the greatest difficulties that beginners
have. Frequently they suppress an e mute which the French
pronounce, and this is a much worse fault than pronounc-
ing one which is generally suppressed. At all events, the
beginner's choice in the matter is rarely a happy one, and he
would be well advised to study this subject thoroughly. It
is impossible to lay down precise rules, as the elision of the
sound often depends on taste, on the degree of emotion, or
on the rate of utterance, and varies not only in the speech
of different people, but frequently in that of the same in-
dividual. There are certain definite principles, however,
which we propose to mention for the student's guidance. It
requires to be emphasized that these principles apply only
to familiar and current conversation, inasmuch as the
slower or more emphatic or more elevated the speech is, the
less does elision take place. Learners would do well to
elide little until rapidity of speech makes elision natural.
'^ 176. The general principle is that the e mute sound is
elided in all cases, except where its elision would bring
three or more consonants together. This is known as the
Law of the Three Consonants. The name thus given is
perhaps too strict, as the principle is hardly a ' law,' but
it at least amounts to a very general tendency manifesting
itself in the great majority of cases. Thus, we have les
ch{e)vaux [le Jvo], la cite d(e) Paris [la site d Pari], but plusieurs
chevaux [plyzjoeir Javo], la ville de Paris [la vil da Pari]. Note"
that, as we are dealing with the language phonetically, the
student must understand by * consonants ' pronounced
ones.
177. The principle may be more clearly stated under two
rules :
(i) When an e mute is separated from the pre-
WORDS IN COMBINATION
151
ceding vowel
elided, as :
by one consonant only, it is always
La p(e)tite [la ptit].
Le ch(e)val [I9 Jval].
Tout 1(e) monde [tu 1 moid].
Rach(e)ter [rajte].
J'aim(e)rai [38mre].
Tu d(e)mandes [ty dmaid].
Au r(e)voir [0 rvwair].
C'est lui qui 1(e) dit [ss Iqi ki 1
di].
Tous mes r(e)grets [tu me rgre].
II n'a pas d(e) scrupules [il na pa
d skrypyl].
Point d(e) viande [pw£ d vjctid].
C'est c(e) qui fait mal [ss ski fs
mal].
Va t(e) coucher [va t kuje].
C'est c(e)pendant droit [ss spddci
drwa].
It is in virtue of this rule that the e mute sound is elided
at the end of certain words, as un{e), dam{e), homm(e) , fair{e) ,
collin{e), etc. Further examples of the rule are here given :
Brac(e)let, pel(e)rin, la p(e)louse, sur(e)te, mull(e)tier,
bonn(e)tier, souv(e)nir, viv(e)ment, even(e)ment^ chaud(e)-
ment, om(e)lette, bull(e)tin, el(e)ver, pal(e)tot, mad(e)moiselle,
a d(e)mi, je donn(e)rai, tu r(e)tournes, le bouquet d(e)
prim(e)veres, le mari d(e) madame^ est-c(e) vrai, vous m(e)
permettez, rue d(e) la Paix^ le roi d(e) Gr^ce, on n(e) veut
pas, dans 1(e) bois, donnez-moi 1(e) specimen.
(2) When e mute is separated from the preceding
vowel by two or more consonants {i.e. two or more
different ones), it is generally sounded,^ as :
La riviere de diamants [la
rivjeir d9 djama].
Un OS de poulet [den os d9 puis].
Une chaise de salon [yn Jsiz d9
salo].
Un soleil levant [de solsij bva].
Ordre du jour [ordr9 dy 3uir].
Quelquefois [kElk9fwa].
Ventrebleu [vdtr9bl0].
Je rentrerai [39 ratr9re].
Entreprise [atr9priiz].
Tristement [trist9ma].
Arbre vert [arbr9 vsir].
La femme de chambre [la fam
d9jaibr].
An e which is elided under rule (i) is of course regarded as
non-existent, so that we have two consonants together in such
cases as un(e) demoiselle, un{e) fenetre, on n{e) le voit pas, il
mang(e) le pain, etc.
^ In careless or rapid speech there are divergencies from this rule,
especially in the case of final e mute see §§ 186, 187).
152 MANUAL OF FRENCH PRONUNCIATION
Further examples of rule (2) are here given for practice :
Premier, aprete, justement, exactement, porte-croix, porte-
plume, sobrement, parlement, gouvernement, puisque tu
veux, presque tous, quelque chose, cett(e) fenetre, ch^r(e)
petite, Paul refuse, un verr(e) de vin, une longu(e) semaine,
un porte-bonheur, table d'hote, un artiste-peintre.
178. As a sounded e mute is equal in value to an ordinary
vowel, it is evident that the above rules may be applied to
several syllables in succession, and thus extended indefinitely.
For example :
Qu'est-c(e) que j(e) te disais ?
Nous n(e) te 1(e) demandons pas.
Qa n(e) te r(e)garde pas.
II se r(e)pose pr^s d(e) la ch(e)-
minee.
Vous n(e) le d(e)venez pas.
kes ka 3 t9 dize ?
nu n t9 1 damado pa.
sa n t9 rgarda pa.
il s9 rpoiz pre d la /mine.
vu n I9 dv9ne pa.
-J The ' Law ' of the Three Consonants, however, as we have
stated, is not absolute under all conditions. It has ex-
ceptions, and there are also cases where it is inapplicable,
and where other treatment is required. We proceed in the
rest of this chapter to explain these special cases.
I. Initial Syllables
179. If e mute occurs in the first, or first and second
syllables of a phrase (as in ]e nCen vats, je ne sais pas),
its elision depends on the nature of the consonants accom-
panying it :
(i) When it occurs in the first syllable only, it is
generally elided if the preceding consonant is a non-
explosive, but it is sounded if it is an explosive (p, b, t, d,
k, g), as j{e) vous remercie, c{e) n'est pas ga; but, te visite-t-il ?
que pensez-vous ? The reason is that non-explosive consonants
can be uttered without a pause (hence called continues by
the French), while explosives stop suddenly (hence called
momantanees) and require a vowel to sustain them. Further
examples :
WORDS IN COMBINATION
153
Explosives
Te faches-tu [t9 fai/ty] ?
Te trouves-tu bien [t9 truivty
bJE] ?
Que faites-vous [ka fetvuj ?
De pr^s, c'est beau [d9 pre, se
bo].
De ill vient ma peine [d9 la vje
ma pen].
Que m'importe [k9 meport]
Te faut-il ceci [ta fotil S9si] ?
Non-explosives
J(e) t'6coute [3 tekut].
J(e) donne mon temps [3 don mo
ta].
J(e) cherche a plaire [3 JerJ a
pleir].
C(e) papier est a moi [s papje et
a mwa].
N(e) faites pas la belle [n fet pa
la bel].
L(e)vez-vous [Ivevu].
J(e) cr^ve de faim [3 kreiv d9 fe].
At the same time, when clearness is necessary, the e in the
first case should be sounded, and especially when it comes
between two identical consonants, as /e jeite, ce sucre est
brut ; and in the second case, it may be elided before a non-
plosive, as qu{e) voulez-vous ? It is eUded in such words as
p{e)lote, p[e)loton, p{e)louse, p{e)luche, p{e)lure, b{e)lette, etc.,
where a natural group of consonants comes together.
-^ (2) When it occurs in both the first and second syllables,
the * Law * of Three Consonants becomes applicable,
and by virtue of rule (i) of that ' law,' the second e is
elided, as que r{e)gardez-vous ? je n{e) sais Hen, que t[e)nez-vous ?
But note an important exception, viz., that if the first con-
sonant be a non-explosive and the second an explosive, it is
the first e that is elided, as j(e) te trouve bien, c(e) petit gargon.
Further examples :
Second e elided
Que n(e) vas-tu pas [k9 n vaty
pa]?
Que j(e)tez-vous [k9 3tevu] ?
Je n(e) peux pas marcher [39 n
p0 pa marje].
Je 1(e) tiendrai [39 1 tjedre].
Je 1(e) connais [39 1 kone].
Te 1(e) donne-t-il [t9 1 dontil] ?
Te 1(e) rappelles-tu [t9 1 rapelty] ?
Dev(e)nez plus sage [d9vne ply
sai3].
Rec(e)vez ma sympathie [r9sve
ma sepati].
First e elided
J(e) deviens riche [3 d9vje rij].
J(e) te vois jouer [3 t9 vwa 3we].
N(e) te leve pas [n t9 leiv pa].
N(e) te I'ai-je pas dit [n t9 lei3
pa di].
N(e) te fache pas [n t9 faij pa].
C(e) que j(e) dis, c'est vrai [sk9
3 di, se vre],
C(e) que c(e)la signifie, j(e) te 1(e)
dirai [sk9 sla sijiifi, 3 te 1 dire].
J(e) te 1(e) red(e)vrai [3 t9 1
r9dvre].
J(e)te l(e ) propose [3 19 1 propoiz].
m
154 MANUAL OF FRENCH PRONUNCIATION
2. Fixed Groups
^180. It happens that when some groups of two syllables,
each containing an e mute (e.g. je ne) are pronounced in a
certain form initially, they retain this form in the interior
of sentences, where the strict application of the rules would
act otherwise. Thus, as already seen, the group je ne,^ when
occurring initially, is pronounced je n{e), and hence it pre-
serves this form of pronunciation, if at all possible, in all
other positions, as :
ska 39n v0 pa.
8s k9 39n l9 kone pa ?
si 39n p0 partiir, S8 domai3.
si 39m deple isi, 39 partire.
ka 39m fajre, 39I grodre.
3ire Je Iqi, e 39I propozre.
C(e) que je n(e) veux pas.
Est-c(e) que je n(e) le connais
pas ?
Si je n(e) peux partir, c'est
dommage.
The same is true of at least two other groups, je m{e), and
je 1(e), as :
Si je m (e) deplais ici, je partirai.
Quand je m(e) facherai^ je 1(e)
gronderai.
J'irai chez lui, et je 1(e) pro-
poserai.
These three groups referred to are largely fixed ones, for
which a form with [9] in the second syllable is not normal.
When we reflect that the pronoun je, though occurring in
the interior of a sentence, generally commences a new pro-
position or statement, and is thus practically equivalent
to an initial word, we can understand the fixity of the com-
binations. The group je ne, which is the commonest of all,
is so definitely fixed that the form j{e) ne would hardly be
regarded as French. The groups je me and je le occasionally
take the other form in obedience to the general rule, although
the fixed form is the more frequent. We might say, e.g. et j{e)
le proposerai, si j{e) me deplais ici, etc., but the other forms
^ are more usual.
181. There are several other groups of two syllables which
have their origin in the interior of sentences, but which
are nevertheless fairly well fixed, such as que j{e), que
l{e), de l[e), de n[e), te l{e). They owe this form to the fact
WORDS IN COMBINATION
155
that in the large majority of cases they follow a consonant,
being frequent in such phrases as est que j{e) . . ., parce
que j{e) . . ., puisque j{e) . , ., tout c{e) que j{e) . . .,
est-ce que l{e) . . ., etc. The group que j{e) in particular is
well fixed, but the rest often succumb to the general rule
when they come into conflict with it. Examples :
Voulez-vous que j(e) parte ?
II a decide de n(e) pas sortir.
II a ete forc6 de 1(e) punir.
II croit que 1(e) combat est
necessaire.
Je veux te 1(e) demander.
Voulez-vous que j(e) lui ecrive ?
Je veux que 1(e) gargon vienne.
II a tachi de 1(e) sauver.
lis ont choisi de n(e) pas rester.
A-t-on ose te 1(e) dire ?
vulevu k93 part ?
il a deside don pa sortiir.
il a ete forse dal pyniir.
il krwa kol koba 8 nesEsEir.
59 v0 t9l d9made.
vulevu k93 Iqi ekriiv ?
39 v0 k9l garso vjen.
il a tcje d9l sove.
ilz 3 Jwazi d9n pa rsste.
ato oze tgl diir ?
182. A few other groups, such as que d{e), de m{e), etc.,
have a tendency to appear fixed, but are much less so than
those already mentioned. Thus, we may say il a envie de
m{e) plaire, or il a envie d{e) me plaire, either form being
equally good.
"' 183. Note that when two groups come into conflict, the
more stable of the two maintains itself. In this respect
je n{e) is the strongest of all, as :
Voulez-vous qu(e) je n(e) le dise
pas ?
Tu vois qu(e) je n(e) te 1(e)
demande pas.
II croit qu(e) je n(e) te 1(e)
red(e)mand(e)rai pas.
vulevu k 39n I9 diiz pa ?
ty vwa k 39n t9l d9maid pa.
il krwa k 39n tol rodmadre pa.
3. The Prefix RE-
/ 184. When this prefix is preceded by a monosyllable
• containing e mute (such as je, te, me, le, etc.), the mono-
; syllable generally retains its e, and the one in re- is elided, as :
156 MANUAL OF FRENCH PRONUNCIATION
Cette eau se r(e)froidit.
Jean se r(e)pose sur ses lauriers.
Cet ecolier se r(e)lache.
J'ai tente de r(e)commencer.
L'affaire ne te r(e)garde pas.
C'est le r(e)venu foncier.
Je r(e)tourne ce matin.
set 0 S9 rfrwadi.
3a S9 rpoiz syr se brje.
set akolje s9 rlaij.
30 tate d9 rkomase.
lafeir n9 t9 rgard pa.
se I9 rv9ny fosje.
39 rtum S9 mate.
(i
If, however, a fixed group of a fairly strong type, such as
^e ne, etc., immediately precedes re-, the group retains its
form, with the result that the e of re- remains, as :
Tu croyais qu(e) je n(e) revien-
drais pas.
J'ai dit qu(e) je n(e) retour-
nerais pas.
II choisit de n(e) ret(e)nir ni
Tun ni I'autre.
ty krwaje k 39n r9vjedre pa.
3e di k 39n raturngre pa
il /wazi dan ratniir ni Ide ni
loitr.
But where the group is less stable, it gives way before re-,
as j{e) me r{e)pose, j{e) le r{e)tiendrai, etc.
f.
4. The [5] Sound followed by E 'Mute'
185. It happens that three consonants, or four if the last
be r or /, can easily be pronounced together if the second one
1 is s (or c * soft '), as in abstinence, ahstrait, expirer (ksp),
\ rair stupide, etc. This being so, there is no need for an
e mute after the s sound, and it is consequently elided, as
U_s{e)ra, Arthur s{e) moque de moi, je Vai dit parc{e) que
c'etait vrai, Us comprennent c{e) que c'est, il s{e) trouve bien.
/ If there be an e mute after the first of the three consonants,
it is elided also, under rule (i) of the general ' law,' so that
in this case too we have [s] standing between two or three
consonants, as il n'y a pas d{e) s(e)cours, je n'ai pas d{e)
s{e)cret, Henri n(e) s{e) lamente pas. In ordinary con versa- \
tion, indeed, the e of se or ce is never retained except before /
the prefix re-. Further examples : ^
Ces deux amis n(e) s(e) con-
naissent pas.
Nous travaillons mieux qu(e)
c(e)la.
se d0z ami n s koneis pa.
nu travajo mj0 k sla.
WORDS IN COMBINATION 157
il sa3i d slqi ki a fe 1 mal.
Iwi n s dut pa d ska vu fet.
3 v0 parle d sk9 nu savo^
II s'agit d(e) c(e)lui qui a fait le
mal.
Louis n(e) s(e) doute pas d(e)
c(e) que vous faites.
Je veux parler d(e) c(e) que nous
Savons.
It will be noticed that c{e) que is a fixed group, for even
though it follows a vowel instead of a consonant (as je sais
c{e) que vous dites), the first e is elided according to the
general rule. The pronunciation ce qu{e) which is sometimes
heard is not good French.
5. Final E Mute, after a Consonantal Group
186. When final e mute is preceded by a group of two con-
sonants (as in quatre, feuple, reste, etc.), it may or may not
be elided as the speaker chooses. The following observations
will make the matter clear :
If the last consonant is a liquid (r or 1), there are
three courses open :
(i) In careful speech the e mute is sounded in all cases,
in accordance with the general rule, unless before
a vowel, as pauvre France, autres temps auires moetirs,
la noble Belgique, faire table rase, un obstacle
gigantesque.
In this connexion, beginners should guard against the e
mute sound creeping in before the liquid instead of after it,
as it always does in English words of the same kind (e.g.
theatre, possible, etc.). Oracle, ancre, centre, etc., must not
be pronounced [oraikal, aikor, saitgr], but [oraikla, aikra,
saitrg], not the least breath-sound being heard between the
two consonants.
(2) In less careful conversation, the e mute may be
elided, and the liquid becomes reduced, completely
losing its voice after a voiceless consonant, and partially
losing it in other cases, as in the words quatre [katr],
poudre [pudf], peuple [poepl], seigle [ssgl]. In such
cases the liquid is too weak to form a separate syllable,
and becomes part of the previous one. Examples :
158 MANUAL OF FRENCH PRONUNCIATION
lis ont la fievr(e) politique.
Cela peut lui etr(e) bien utile.
II montr(e) du courage.
La pluie gonfl(e) ces torrents.
ilz 5 la fjeivr politik.
sla p0 Iqi 81 tr bjsn ytil.
il moitr dy kurai3.
la plqi gofl se tord.
This pronunciation may be easy when the word is final,
or before a pause (such as the end of a stress-group),
and is adopted frequently in such cases, but it is rather diffi-
cult when followed at once by a consonant, and hence
it is generally avoided in such connexions.
(3) Most people, therefore, at least in their ordinary
careless speech, simplify the pronunciation im-
mediately before a consonant by totally suppressing
both the liquid and the e mute, as aut(re) chose
[oit /oiz], le pauv{re) gargon [I9 poiv garso], rend(re)
service [raid servis], un meub{le) d' occasion [ce moeb
dokozjo], une tab{le) d' acajou [yn tab daka3u], im-
possib{le) de le faire [eposib dal feir]. The [r] in such
cases disappears more easily than [1]. Indeed, it
disappears sometimes when the word is final, as ils
sont quat{re). Even careful speakers, when talking
rapidly, suppress it in the familiar words noire, voire,
quaire, as noi{re) ami, not{re) iahle, quat{re) personnes ^ ;
and it is always dropped in compound expressions like
un maU(re) d'hoiel [met dotel]. But the liquids reappear
before a vowel, at least in correct French, as un autre
affaire, un meuhle uiile, impossible aujourd'hui.
These simplifications are regarded as excessive and even
slovenly by many excellent teachers, and certainly they
should only be indulged in when talking with a kind of care-
less freedom. Further examples :
Un mait(re) d'ecole, vot(re) legon, prend(re) I'express,
arb(re) fruitier, une lett(re) de recommandation^ il doit et(re)
puni, apr^s m'et(re) rase, il veut et(re) compris, un triang(le)
rectang(le), rartic(le) du Temps, un obstac(le) formidable.
187. If the last consonant is not a liquid (e.g. resie,
posie, force, etc.), it cannot be suppressed. Before a con-
sonant, the group must either be pronounced entire, along
with the e mute (see § 177 (2)), or the e mute may be eUded
1 Exception must be made in Notre Pere, Notre-Dame, etc., where
reverence maintains the full sound, and also in quatre-vingts.
WORDS IN COMBINATION 159
in those cases where it does not bring a difficult com-
bination of consonants together. The former method is
generally adopted in careful speech, and the latter in quick,
familiar conversation.
In the latter, the bringing together of two explosives would
form a difficult combination, and in such cases the e must
be retained, as un aYtiste-peintY{e), un porie-bonheur, un triste
deuil, il teste dehout. Even where only the first of the two con-
sonants is an explosive, the e is often retained, as is generally
done in the words presque, puisque, jusque, quelque. Thus,
we say jusque Id [3ysk9 la], quelque chose [kelka Joiz], presque
jamais [preska 3am8], etc.
But in all other cases the e may be elided in familiar con-
versation, as tourn{e)-toi, divers(es) mithodes, assist(e)-le. It
may also be elided before a pause, such as the end of a
well-defined stress-group, as tu es tnst{e) ce soir, elle est
mort{e) sans souffrir, cette voyelle disparait presqu{e) dans le
parley. In rapid speech it may even be elided in the middle
of a word, provided the combination of consonants is an easy
one, as je rest{e)rai, j'observ{e)rai, etc. Further examples :
Cast just(e) qu'il meure, la post(e) s'est trompee, le rest(e)
n'importe pas, cela port(e) bonheur, il nargu(e) ses ennemis,
une plan^te perc(e) la nue, un artist(e) lyrique, appart(e)ment,
fourb(e)rie, etourd(e)rie, lampist(e)rie, etc.
6. Fixed Words
188. There are a few Words which contain two or more
e mute syllables in succession, and of which the interior form
is fixed. Such are ensev{e)lir [asavliir], ichev{e)U [e/9vle],
ressem{e)ler [rasamle], redev{e)nir [radavniir]. In all four
cases, it wiU be noticed, it is the last e that is elided, owing
to the fact that it precedes the stressed syllable. The first
two words never change their form, but the other two lose their
second e when the third is replaced by an ordinary vowel, as :
Ress(e)melle mes chaussures.
Je n(e) ress(e)melle pas las
chaussuras.
II rad(e)viant malade.
Je n(e) red(a)vians pas pauvre.
rasmel me Josyir.
39n rasmsl pa la Josyir.
il radvje malad.
39n radvje pa poivr.
i6o MANUAL OF FRENCH PRONUNCIATION
Under all other conditions they retain the form fixed in
the infinitive, as :
Fait(es) ressem(e)ler les bottines.
II fait ressem(e)ler les bottines.
Je n(e) redev(e)nais pas malade.
Que voulez-vous r(e)dev(e)nir ?
fet r9S9mle le botin.
il fs r9s9mle le botin.
39n r9d9vns pa malad.
k9 vulevu rd9vniir ?
Verbs in -eter (bequeter, caqueter, etc.) suppress their two
e's in the future and conditional forms (except in ist and
2nd plural conditional, for which see next section), as
il bequ(e)t{e)ra, il caqu[e)t{e)rait, ils coll(e)t{e)ront, etc.
7. E Mute pronounced
189. There are several conditions under which e mute
is never elided, even in the most familiar speech :
1. Before ri- or li- in those cases where the i is always
pronounced as [j] and regarded as a consonant. This is the
case in rien, and in the first and second plural of the condi-
tional of verbs, as en moins de rien [d mwe d9 rje], il ne vous
demande rien [D. n9 vu dma!d9 rjg], nous serions heureux [nu
S9rj5z oer0], vous marcheriez a voire ruine [vu marj9rje a votrg
rqin], nous causerions ici [nu koz9rj3 isi]. Similarly we have
appelions [ap9lj5], appeliez [ap9lje], Richelieu [rij9lj0], and
even ce roi [s9 rwa], ce ruisseau [s9 rqisb]. The reason is
that a liquid (r or /) cannot maintain itself in a group of three
real consonants unless it is first or last, not second.
2. Before the terminations -nier, -Her in nouns.
Thus, centenier [sat9nje], batelier [bat9lje], chandelier [Jad9lje],
bachelier [baj9lje], un denier [de d9nje].
3. Before h 'aspirate.' As explained (§166), the h 'aspirate*
is no longer pronounced in French, but it prevents liaison
or elision before it, exactly as if it were a consonant, as une
halle [yn9 al], dames de halles [dam d9 al], une hausse [yng
01s], le hamac [I9 amak], un triste here [de trist9 sir], c'est
une honte d'agir ainsi [set yn9 oit da3iir £si], quels trisies heros
que ces hussar ds ! [kel trist9 ero k9 se ysair !]. Similarly,
before un (noun of number) huit, onze, as le un, il ignore que
onze et deux font treize (see § 170).
4. In the pronoun le after an imperative, even though
followed by a word beginning with a vowel, as dis-le [diloe],
WORDS IN COMBINATION i6i
faisons-le maintenant [fazoloe metna], rendez-le aimable [mdeloe
smabl]. In all such cases the e mute is stressed, being the
final syllable, and consequently it is not only retained
but really becomes [ce] (§ 70) . In conditions where it is
unstressed, it follows the general rule, as preiez-l{e)-moi
[pretelmwa], mets-l[e)-moi sur la table [mslmwa syr la tabl],
in which cases the stress falls on moi and not on le.
190. Before concluding this chapter, the fact needs em-
phasizing that the more elevated or more sustained the
tone of the speech is, the more is the e mute pronounced.
For instance, clearness may necessitate its pronunciation,
as tu devrais te teindre les cheveux ; or contrast, as ce n'est
pas apres d(e)main, c'est domain. In slow and emphatic
speech, almost every e mute is sounded. One might say in
rapid, familiar conversation, allez, j(e) vous r{e)fuse, but the
refusal is more emphatic or tragic if one says [ale, 39 vu rafyiz].
A careless priest will speak of les commayidements d{e) Dieu,
but another with more reverence will say les commandements
de Dieu.
191. In verse, all e mute vowels in the interior of the
lines are pronounced, and in music there are separate notes
provided for them. This is due to the fact that in French
verse, unlike English, the rhythm does not depend solely upon
the regular distribution of accented {i.e. stressed) syllables,
but also upon the number of sounded syllables being equal in
each corresponding line, and for this purpose the e mute is
always reckoned as one, except before a vowel or h ' mute ' or
at the end of a line. Some people, yielding to the analogy
of prose, suppress some of the e mute vowels, thus curtailing
the proper number of syllables in each line. This may suit
the less elevated styles of poetry, and indeed many popular
ditties are intended to be recited or sung without a strict
adherence to the e mute, but in elevated verse at least every
syllable should be sounded wherever possible, so long as the
form of the language is not abused. In classic and romantic
verse particularly, it is a mistake to omit any e mute syllable.
It should be remembered that the basis of such verse is
artificial and archaic, and that to suppress any of the syllables
is really to falsify the measure.
192. It remains to be said that in popular speech e mute
is often inserted where it does not occur in the ordinary
i62 MANUAL OF FRENCH PRONUNCIATION
spelling. It appears at the end of consonantal groups, or
in their interior. Thus, we hear un ours blanc [oen urs9 bla],
fare de triomphe [larka d trio if], c'est Max [s8 maks9], Felix
Faure [feliksafoir], I'est de la France [lesta d la frdis], Ernest
Blanc [srnesta bla], lorsque [brsaka], expres [eksaprs]. Culti-
vated French speakers avoid such insertions of e, regarding
them as due to slovenliness or lack of education.
EXERCISES
I. The following sentences should be studied and carefully
practised :
I. Je n(e) peux m(e) souv(e)nir de c(e) nom. 2. Je n(e) te
1(e) demand(e) pas. 3. Crois-tu que j(e) me r(e)pens ?
4. Comment puis-j(e) me rappeler 1(e) visag(e) de c(e) gargon ?
5. Qu'est-c(e) que j(e) te disais ? 6. N(e) te fach(e) pas d(e)
c(e) que je n(e) te 1(e) dis pas. 7. Je s(e)rai Rustan, et je
n(e) le s(e)rai pas. 8. Qu'est-c(e) que tout c(e) que j(e) vois ?
9. II s(e)ra bien force d(e) s(e) taire. 10. C(e) n'est rien d(e)
c(e) que vous pensez. 11. De c(e) que je n(e) le d(e)mand(e)
pas, n(e) concluez pas que je n(e) le veux pas. 12. Tu t(e)
lament(es) de c(e) que je n(e) te le r(e)mets pas.
II. Read the following aloud, with special attention to the
elision of e mute :
Je suis perdu, on m'a derobe mon argent. Qui peut-ce
etre ? Qu'est-il devenu ? Ou se cache-t-il ? Que ferai-je
pour le trouver ? Ou ne pas courir ? Qui est-ce ? J'ignore
ce que je fais. Helas ! mon pauvre argent, on m'a prive
de toi ! Et puisque tu m'es enleve, je n'ai plus que faire au
monde. C'en est fait, je me meurs. II faut, qui que ce soit
qui ait fait le coup, qu'avec beaucoup de soin on ait epie
I'heure, et on a choisi justement le temps que je parlais a mon
traitre de fils. Je veux aller faire donner la question a toute
ma maison. Que de gens assembles ! Je ne jette mes regards
sur personne qui ne me donne des soupgons, et tout me semble
mon voleur. Eh ? De quoi est-ce qu'on parle la ? De celui
qui m'a derobe ? lis me regardent tous et se mettent a rire.
Je veux faire pendre tout le monde ; et si je ne retrouve mon
argent, je me pendrai moi-meme apres.
MoLiERE, UAvarej condensed
CHAPTER XXXII
TONIC OR RHYTHMIC ACCENT
(accent d'intensite)
193. By * accent ' we here mean the special stress that is
given to a particular syllable when it is uttered with more
energy or force than the others and sounds somewhat louder.
For example, in the English word tendency, the stress is on
the first syllable, in ability on the second, and in comprehend
on the third. There is a large amount of such stress in English
and other European languages, but in French there is a
remarkable absence of it. There are of course changes in
pitch, and almost infinite shades of voice due to syllables
being strong or weak, clear- vowelled or dull, but these changes
are not necessarily connected with stress. The fact is that
all syllables in French are almost equally stressed, so
that there is a kind of monotonous uniformity in their utter-
ance. In listening to a French speaker one cannot help
noticing this even rhythm, caused by every syllable being
uttered with almost equal force and distinctness. As long
ago as 1567, Plantin, in his Dialogues frangoys, stated that
the French language did not recognize any accent, and three
hundred years later, Nisard declared that " each syllable
has the same accent in French."
194. At the same time there is a slight stress, generally
tetmed the Tonic or Rhythmic Accent, which occurs as de-
scribed in this chapter. Students must beware of exaggerating
it : it is by no means so strong as it is in English. Frequently,
indeed, it is so slight that it is almost imperceptible.
(i) As a general rule, every word when isolated has
the accent on the last sounded syllable (unless this is an
e mute), as bonte, frangaiis, colossal, bourgeoisie, epee, article ;
and this is the case even with words which have been borrowed
from other languages where the stress has been generally
on a previous syllable, as revolver, jockey, gratis, cortes,
163
i64 MANUAL OF FRENCH PRONUNCIATION
Luther, Eldorado, If a previous syllable contains a long vowel
(e.g. baron, beaucoup, fdche, heros), this lessens the stress
on the final syllable, making it almost imperceptible. Many
French words are thus in a state of almost perfect equilibrium.
This accent on the last syllable of isolated words is in
most cases a survival of the Latin forms. In Latin, words
generally had the stress (' ictus ') on the second last syllable,
but as their last syllable has now disappeared in French, the
words end on the stressed one. Thus the Latin accentum,
with the stress on the second last syllable, has become accent
in French, mercsUum has become marche, s-nd fenestra fenetre.
Similarly in Spanish, Portuguese, Italian, and the other
Romance languages (which are derived from Latin), the
accent still f^dls on the same syllable as in Latin, but in
these languages the strong syllable is often followed by
other syllables, as Italian amore (Fr. amour), Spanish co^o
(Fr. coude), Portuguese hera (Fr. lierre). In this respect
the Romance languages differ from the English and others in
the Teutonic group, which usually have the accent on the first
or root syllable. The difference so far as French and English
are concerned, will be evident from the following comparison :
Latin
French
English
Gloriosum
glorieux
glorious
Advocatus
avoue
advocate
Bonitatem
bonte
bounty
Castellum
chateau
castle
Peregrinum
p^lerin
pilgrim
Cellarium
cellier
cellar
Angelum
ange
angel
}5. Where simi
lar words are comi
non to English and
French, care should therefore be taken that the stress is not
placed where it is in English. Beginners are apt to do this.
Such cases, e.g., as the following should be noticed, as they
are very numerous :
English
French
English
French
History
histoire
money
monnaie
Literature
litterature
possible
possible
Government
gouvernement
reason
raison
Companion
compagnon
fashion
fagon
bachelor
bachelier
dozen
douzaine
cushion
coussin
banner
banni^re
WORDS IN COMBINATION 165
196. (2) Words, however, do not always occur isolated,
except in dictionaries and under special conditions. We
have rather to deal with words grouped together into spoken
language. Here the tonic accent does not fall on the last
syllable of every word. This would altogether spoil the
proper flow. It falls upon the last sounded syllable of a
word or group of words expressing within itself a single idea
or concept or picture. This group may contain several
words, but if it expresses only one idea, the whole group is
treated as one long word, and all the sjdlables in it are uttered
with exactly the same force except the last, which has a
very slight stress. It is the recurrence at intervals more or
less regular of these stressed syllables that constitutes the
rhythm of the language.
197. How is such a group known ? It may generally be
known by the fact that it can be replaced by a single
word (a single thought- element), if not in French, at least
in some other language. Thus in the sentence
// y avail une fois une jeune fille qui refusait d' alter d Vkole,
there are five such groups, and consequently five slight
stresses, as follows :
(i) II y avail, which simply indicates exislence in Ihe past
and can be expressed in Latin by the one word eral.
(2) une fois, which localizes the existence in time, and
might be otherwise written in French by aulrefois. If instead
of this we were to say en mil neuf cenl dix-neuf, there would
be six words in place of one, but these six would still constitute
a single group, and only the last syllable neuf would be
stressed.
{3) une jeune fille, which also contains a single thought-
element, for it designates a particular category of human
beings, like enfanl, homme, pelil gargon, vieille femme, and
might be translated by the Latin word puella. The ex-
pression un fille jeune, on the other hand, would constitute
two groups, for it refers to an individual of the class fille,
who possesses a particular quality (jeune),
(4) qui refusail, which might easily be replaced in French
by refusant.
(5) d'aller a I'ecole, which is practically equivalent to
i66 MANUAL OF FRENCH PRONUNCIATION
d'etudier, the word alley being here a mere grammatical
complement. If, however, we were to say d'aller a I'ecole,
a rSglise, ou au museum, the. expression d'aller would form
a group by itself, because the idea of going becomes a
separate one, connecting itself with other objects than the
school.
The whole sentence has thus five easily defined groups, the
various syllables of which receive equal force and distinctness,
except the last of each which is slightly stressed, as :
II y avait | une fois | une jeune fille | qui refusait | d'aller k I'ecole.
198. It will thus be seen that though a word when isolated
has an accent on the last syllable, it may lose that accent in
a group, unless it is the last word in it ; for the accent belongs
not to the word but to the group. Hence we have :
Avant, but avant-coureur
Lieutenant, ,, lieutenant-colonel
Savoir,
Nous avons,
II laissa,
Les champs,
Le jardin,
II aime.
Comment,
Pendant,
savoir-vivre
nous avons eu
il laissa tomber
les Champs-Elysees
le Jardin des Plantes
aime-t-il
comment allez-vous
pendant leur entretien
199. These groups, known as stress-groups, to which we
refer in this chapter, must not be confused with breath-
groups, or clauses, which depend on the exigencies of breath,
or on the requirements of expression. Sometimes, no doubt,
the one may coincide with the other, but as a rule, breath-
groups are generally long enough to contain several stress-
groups. Thus, the phrase, honjour, mon cher ami, forms a
single breath-group, but it contains two stress-groups.
200. The following cases of non-accentuation should be
noted :
(i) There are of course a large number of little words,
such as pronouns, prepositions, conjunctions, etc., which
grammarians call proclitics or enclitics, and which by their
very nature are alv/ays unstressed. They may, either alone
WORDS IN COMBINATION 167
or in combination, express an idea, but they cannot constitute
a stress-group. Thus, in the sentence
Dites-moi | ce que vous ne voulez pas
there are only two accented or stressed syllables, moi and pas.
Theoretically, there should be an accent on ce que, which
contains a well-defined thought (equivalent to la chose que),
but under ordinary circumstances the weak e mute vowel
cannot take an accent, and indeed is not pronounced at all
except where necessary. All such words fall into the
adjoining group. If they are. not final, or if there is no
pause after them, they fall into the group which follows,
never that which precedes ; if final, they fall of course into
the preceding group. Thus :
Qu'est I -ce que le maitre | a demande | a ses ecoliers ?
Ce n'est rien | de ce que vous pensez.
L'homme, | qui est egoiste, | abuse | du plus faible.
II y en avait | qui, au lieu de pain, | mangeaient | des pommes.
Un peintre | qui, des le matin, | travaille | a ses tableaux.
Qu'est-c(e) ? | Que sais-j(e) ?
In some of these examples it will be noticed that even the
comma does not always divide one group from another.
Provided there be no pause after the relative, it becomes a
proclitic word, which cannot take the stress, and hence the
division takes place before and not after it.
201. (2) Many adjectives, adverbs, and similar words,
which are inserted within a stress-group to express shades
of meaning", do not usually receive the accent, although
theoretically they should have it. Thus, il a Men iravailley
il a beaucoup travaille, il a parfaitement chante, les plus
profonds mysteres, un grand tableau , de beaux infants, un
excellent ami, les moindres offlcievs, etc.
Such words, however, frequently receive the emphatic
accent or stress, especially in elevated or sustained speech,
but this is a different matter, referred to in Chapter XXXIV.
202. (3) When a stress-group, complete in itself (e.g. il
parlait), is followed by a final monosyllabic word (as in il
parlait bien), the group usually loses the accent (Disaccentua-
tion), while the monosyllable retains it. In other words,
i68 MANUAL OF FRENCH PRONUNCIATION
the group extends itself to include the monosyllable, as il
parlait bien. Hence we have un garc^on diligent, un gargon
/labile, but un gargon bon. Similarly, we have :
Vous le voyez ? and Vous le voyez done ?
Un grand bruit, „ Un grand bruit d'hommes.
D'une voix,
C'est moi,
L'ami d 'Alfred,
Le roi Guillaume,
Une femme charmante.
La misere terrible,
D'une voix claire.
C'est moi-meme.
L'ami d(e) Pierre.
Le roi George.
Une femme bonne.
La misere noire.
If, however, the idea contained in the monosyllable be a
distinct one, requiring to be put as it were in relief, the
group retains its accent, and the monosyllable also. In this
case there are two groups, generally with a slight pause
between. Thus, elle chantait bien (only one group) refers
solely to the action of good singing, and thus contains a
single idea, whereas elle chantaii bien {two groups) refers
to the action of singing, and in addition indicates that it
was well done, and it thus contains two distinct ideas.
203. While the above rules on the rhythmic or tonic accent
may be regarded as almost universally observed, there are
variations from them among different speakers. Indeed,
probably no two Frenchmen would speak with the stresses at
exactly the same places in a sentence. It may be said that in
elevated and sustained speech, and above all in verse,
stress-groups are generally shorter than in familiar or rapid
conversation, in the sense that adverbs, adjectives, etc.,
which take an accent in the former, may not receive it in the
latter. Such expressions as de terribles combats, la patrie
sacree, etc., might take two accents in elevated speech, but
only one in ordinary conversation. The rhythmic accent
may thus be suppressed or added according to the nature
of the speech. But there is a limit to these variations, and
the student who wishes a correct accent should practise the
division of sentences into * idea-groups,' and aim at monoton-
ous uniformity of aU the syllables, with a very slight stress
on the last one in each group. In other words, he should
give exactly the same force to each syUable in a group, except
the last, which should be very slightly stronger. Those who
WORDS IN COMBINATION 169
are completely under the influence of English accentuation'
may find this difficult at first. Some have found it helpful
to accompany each syllable with a tap on the table, taking
care that the taps are quite regular in time, and all of them
equal in force except the last of each group. Whether this
hint is adopted or not, it is advisable that the speaking or
reading should be done slowly at first, so that each sounded
syllable may receive its full value. It is a mistake to aim
at speed before correctness has been acquired.
We repeat that the accent should only be a slight one.
Some learners have a tendency to make it far too energetic,
which makes the speech jerky and disagreeable, and puts
the accented syllables in exaggerated relief.
The following extract will furnish abundant examples of
the rhythmic accent. A vertical line has been put after
each stress-group. The syllable which precedes a line is
always accented :
La Mer
J'etais arrive | le soir, | avec mes parents^ | dans un village |
de la cote saintongeaise/ | dans une maison | de pecheurs, j
louee I pour la saison des bains. ^ | Je savais | que nous
etions venus ^ la | pour une chose | qui s'appelait la mer,* |
mais je ne I'avais pas encore vue | (une ligne de dunes ^ ] me
la cachait | a cause de ma tres ^ petite taille)^ | et j'etais dans
une extreme ^ impatience | de la connaitre. | Apr^s le diner ^ |
done, I alatombee | delanuit, [ jem'echappai^seul | dehors. |
L'air ^^ | vif, | apre, | sentait je ne sais quoi | d'inconnu, | et
^ La c6te saintongeaise, single idea or concept,
2 Pour la saison des bains, single idea, practically equivalent to pour
VeU.
3 Disaccentuation of venus before lb,.
* Qui s'appelait la mer is here a single idea, qui s'appelait being a
mere rhetorical addition. If s'appelait be accented, there would be two
groups, implying unnecessarily tliat the thing had a name as well as
that this name was the sea.
^ Une ligne de dunes, single idea, equivalent to des dunes.
^ Tres may take the emphatic accent (see chapter xxxiv.).
' ExtrSme may take the emphatic accent.
^ There is no disaccentuation here of diner before done, as the latter
is meant to form a separate idea with a slight pause before it.
* Disaccentuation of nt'dchappai before seul.
^" No disaccentuation of Pair, as the adjectives following are intended
to be set in relief as distinct ideas.
170 MANUAL OF FRENCH PRONUNCIATION
un bruit | singulier, | a la fois faible | et immense, | se faisait
entendre | derriere ^ les petites montagnes ^ | de sable, |
auxquelles un sentier | conduisait. | Tout m'effrayait, ^ | ce
bout I de sentier | inconnu, | ce crepuscule | tombant * d'un
ciel ouvert, | et aussi | la solitude | de ce coin | de village. |
. . . Cependant, | arme | d'une de ces grandes ^ resolutions |
subites, I comme les bebes | les plus timides | en prennent
quelquefois, | je partis | d'un pas ferme. | . . .
Puis, I tout a coup | je m'arretai | glace, | frissonnant ^ |
de peur. | Devant moi | quelque chose | apparaissait, |
quelque chose | de sombre | et de bruissant | avait surgi | de
tons les cotes | en meme temps, | et qui semblait ne pas
finir, '^ I une etendue en^ mouvement | qui me donnait le
vertige ^ | mortel. | . . . Evidemment | c'etait 9a ; | pas ^
une minute | d 'hesitation, | ni meme | d'etonnement | que ce
futainsi, | non, | rien | que de I'epouvante ; | je reconnaissais |
et je tremblais. | C'etait d'un vert | obscur, | presque noir ; |
5a semblait instable,^^ | perfide, | engloutissant ; | 9a remuait |
et ga se demenait | partout | a la fois, | avec un air | de
mechancete | sinistre. | Au-dessus | s'etendait un ciel ^^ |
tout d'une pi^ce, | d'un gris | fonce, | comme un manteau ^^
lourd. I Tris ^^ loin, | tr^s ^^ loin | seulement, | a d'inap-
^ Derridre is an unaccented proclitic.
2 Les petites montagnes, single idea, equivalent to les collines.
3 Tout m'effrayait, single idea, almost equivalent to fetais efjvayL
Tout may take an emphatic accent, however.
^ Tombant is here a mere rhetorical expression, forming part of the
main idea, and is better left unaccented.
^ Grandes may take the emphatic accent.
^ Frissonant may take the emphatic accent (on first syllable), as
well as the rhythmic one (on last syllable), thus becoming frissonant.
If a word of more than one syllable requires an emphatic accent, this
is generally placed as here on the first syllable beginning with a
consonant (chapter xxxiv.) .
^ Qui semblait ne pas finir, single idea, almost equivalent to qui ne
finissait pas.
* Qui me donnait le vertige, single idea, equivalent to qui m'dtour-
dissait.
^ Pas may take the emphatic accent.
^^ Qa semblait instable, single idea, semblait being here a mere
copulative verb, unaccented.
^^ S'etendait is here a mere rhetorical expression, forming part of the
idea contained in ciel.
12 Disaccentuation of manteau before lourd.
^^ Trds may take the emphatic accent.
WORDS IN COMBINATION 171
preciables | profondeurs | d'horizon, | on apercevait | une
dechirure, | un jour | entre le ciel | et les eaux, | une longue
fente ^ | vide, | d'une claire | paleur jaune. ^ | . . .
Pour la reconnaitre | ainsi, | la mer, | I'avais-je | deja vue ? |
Peut-etre, | inconsciemment, | lorsque vers I'age | de cinq | ou
six I mois, | on m'avait emmene | dans I'ile, | chez une
grand'tante, | soeur | de ma grand'm^re. | Ou bien | avait-
elle ete | si souvent regard ee | par mes ancetres | marins |
que j'etais ne ay ant deja^ | dans la tete | un reflet | confus
I de son immensite ? |
Nous restames | un moment | Tun devant I'autre, | moi |
fascine par elle. | Des* cette premiere entrevue, | sans
doute, I j'avais I'insaisissable | pressentiment | qu'elle finirait
un jour I par me prendre, | malgre toutes ^ mes hesitations, |
malgre toutes ^ les volontes | qui essayeraient | de me
retenir. | . . . Ce que j'eprouvais | en sa presence | etait non
seulement | de la frayeur, | mais surtout | une tristesse | sans
nom, I une impression | de solitude | desolee, | d 'abandon, |
d'exil. I . . . Et je repartis | en courant, | la figure | tr^s ^
bouleversee, | je pense, | et les cheveux | tourmentes | par
le vent, | avec une hate | extreme | d'arriver | aupr^s de ma
mere, | de me serrer ^ | contre elle, | de me faire consoler |
de mille angoisses | anticipees, | inexpressibles, | qui m'avaient
etreint la cceur ^ | a la vue | de ces grandes ^ etendues |
vertes | et profondes. | Pierre Loti
^ There is no disaccentuation of fente before vide, as the latter is
intended to express a distinct idea. The word longue may receive the
emphatic accent, in which case the three words longue fente, vide all
become accented.
2 Disaccentuation of paleur before jaune.
^ This may be regarded as one group, containing only one concept
or picture, with no pause between nd and ayant, and it is better to give
it only one accent.
* D^s may receive the emphatic accent.
5 Toutes may receive the emphatic accent.
® Tris may receive the emphatic accent.
' Serrer may take the emphatic accent (on first syllable), as well as
the rhythmic one (on last syllable) .
^ Qui m'avaient itreint le cceur, single idea.
• Grandes may take the emphatic accent.
CHAPTER XXXIII
LIAISON OR LINKING
204. As already stated, the final consonant in the majority
of French words is not pronounced, so that the words as
uttered end with a vowel. When, however, the next word
begins also with a vowel or h ' mute,' or with one of the three
semi-consonants [w, j, q], the final letter of the first word
is often sounded for the sake of euphony, and becomes the
initial of the next word. This is called liaison or linking.
Thus, the final consonant of the verbal form etait is mute
in il etait grand [il ete gra], but it is sounded in il etaitjimmcnse
[il etet imais], where it becomes the initial of the syllable im-.
Similarly, we say trop^aimahle [trop emabl], prenez^-en
[pranezd], disait^-on [dizeto], il faut^ecrire [il fot ekriir], etc.
This linking prevents hiatus, which generally makes the
utterance difficult and disagreeable. For the same reason,
when the third person singular of the interrogative of a
verb ends with a vowel, the letter t is introduced between
the verb and the pronouns il, elle, on (as a-t-il, parle-t-elle,
dira-t-on), the t being really the survival of the Latin termina-
tion of the verb. On the same principle, we have ne voild-
t-il pas, and the technical term va-et-vient, frequently pro-
nounced va-t-et-vient by sailors.
^ 205. When thus linked to the following word, s and x
become [z],/ becomes [v], d becomes [t], and g becomes [k].
That is to say, the voiceless fricatives (s and /) become voiced,
while the voiced plosives {d and g) become unvoiced. For
example :
s ; Les^aunes [lez oin] (same as les zones), vous^avez [vuz ave],
nos^enjants [noz dfa], un grosjiomme [de groz om], le
tiersu^tat [b tjsrz eta], leurs^amis [loerz ami], lesjyeux
[lez 30]; desjiuttres [dez qitr].
172
WORDS IN COMBINATION 173
X : Deuxjiommes [d0z om], chevaux^arahes [Javoz arab],
choux^aigres [Juz sigr]^ iu peux^attendre [ty p0z ataidr].
Similarly, deuxieme, sixieme, dixieme, etc.
/ ; Neuf is the only word in / that takes liaison, as neuf^ans
[noev a], neujjteures [noev oeir], neufjiommes [noev om].
See chapter xxix.
d : Grandjiomme [grat om], pied^d-terre [pjetatsir], quand^on
viendra [kat 5 vjsdra], il n'a pas froid^aux yeux [il na pa
frwat oz j0].
g ; The g becomes [k] in the adjective long, as in un longjiiver
[de 15k ivEir], un long^espoir [de 15k espwair]. The same is
the case sometimes with the substantives sang and rang,
as sangj,llustre [sdk ilystr], rangjiumain [rak yms] ; but
liaison in the case of these two words is now rare and
regarded as affected. One still sings, however, in the
Marseillaise, '^ Qu'un sang^impur [sdk spyir] abreuve nos
sillons,^' and liaison always takes place in the expression
suer sang^et eau [sqe sdk e 0].
p : The consonant p only takes liaison in trop and beaucoup, as
tropJ,diot [trop idjo], beaucoup^appris [bokup apri]. In
familiar talk there is a tendency to avoid liaison with trop,
but a speaker can nevertheless make it without appearing
pedantic.
Other consonants : devant^elle [dovdt el], un peiit^oiseau [ce
potit wazo], chez^eux [Jez 0], franc^alleu [frdkal0], franc^
archer [frdkarje], franc^itrier [frdk etrie], etc. For n,
indicating nasality, see §101.
206. It should be noted, however, that when a final
consonant is pronounced under all conditions, there
is no liaison properly speaking. The consonant is joined to
the vowel that follows, but it undergoes no change. Thus
we have un os a moelle [den os a mwal], un lis asphodele [de
liis asfodel], des fils obeissants [de fis obeisd], des ours affames
[dez urs afame], ious ont parle [tuis 5 parle], index alphahetique
[edsks alfabetik], codex officiel [kodeks ofisjsl], un chef anglais
[de Jsf dgle], sauf erreur [sof eroeir], sud-est [sydest], sud-ouest
[sydwest], un gong enorme [de goig enorm]. The monosyllable
us [ys], however, generally takes liaison with [z] in the
expression us^et coutumes.
2Sfl, There are limits to liaison : it does not take place in all
cases. As its purpose is that of harmony and the avoidance
174 MANUAL OF FRENCH PRONUNCIATION
of hiatus, it is clear that it should not take place in cases
where it would produce a more disagreeable sound than
its absence would do.^ But apart from this underlying
principle, the main difficulty with learners is to know when
to use it, and when to avoid it. One common rule is that
it should only take place between words that are closely
connected in position.and meaning. Another, which amounts
to the same thing, is that it should only take place when no
pause can be made between the words. Either of these
rules is a good guiding principle, but the matter may be put
better and more scientifically as follows : Liaison as a rule
takes place within the interior of a stress-group, and
not between one stress-group and another. Thus :
Nous avons^ecrit | a trois^amis | aujourd'hui [nuz avoz
ekri a trwoz ami ojurdqi].
Other examples are as follows :
II a couru k franc^etrier [il a kury a frak etrie], but il etait
franc | autrefois.
Un prof ond^esprit [de profot Espri], but son esprit est profond |
et lucide.
Un long^ete [de 15k etc], but le temps est long | a qui attend.
C'est un second^Alexandre [set ce s9got Alsksaidr], but il
est capitaine en second | aujourd'hui.
C'est un heureuxjiomme [set den oer0z om]^ but il est
heureux | et insouciant.
208. This rule, as here stated, is almost absolute in reading,
in declamation, and indeed in all elevated and sustained
speech ; but in familiar and current conversation, the
tendency is to avoid liaison as much as possible, confining it
to grammatical cases, especially those which serve to dis-
tinguish the plural from the singular. The whole matter
indeed is largely one of taste, and speakers vary greatly
according to whether their conversation is elevated or familiar,
slow or rapid, educated or illiterate. It can be said with
assurance, however, that in current conversation the linking
of words in the interior of a stress-group, even when it is
1 For instance, tu as^Ste [ty az ote] , and tu les^as [ty lez a] are both
correct and agreeable to the ear, but tu les^as^ote [ty lez az ote] , with
the close repetition of [z], sounds more disagreeable than the absence
of liaison [ty le a ote], and would be better avoided.
WORDS IN COMBINATION 175
unnecessary, does not strike the French ear as unnatural or
unpleasant (unless it be a pataques ^), whereas the linking
of one stress group to another is generally bad taste,
and denotes a pretentious, affected, or artificial mode of
speech. For this reason, the Swiss pronunciation, which
frequently links the stress-groups, is not regarded as attrac-
tive by Parisians.
209. From what we have just said, it will be understood
that it is impossible to give rigid rules in regard to liaison.
But on the main principle that it should take place principally
within stress-groups and not between them, the following
may be regarded as the chief cases where it should be em-
ployed, even in current conversation :
I. Articles, pronouns, possessive and demonstrative
adjectives, prepositions (except selon ^), conjunctions,
as well as adverbs (when fairly short), and similar accessory
words are always linked to the following word when they
govern or qualify it. Thus :
Les_oiseaux^ vous^ecrivez, nous^aimons, ils^offrent, elles^
envoient, on^entre, mon^argent, ces^edifices, leurs^enfants,
quels_idiots, toutjiomme, tout^est fini,^ d'autres^enfants,
pendant^un jour, sans_amis, dans^une heure, chez^eux,
sous^une lampe, en_abondance, en^ecoutant, en_y allant,
bien^aimable, moins_agr cable, plusjionnete, dont^il est,
^ A pataques is a wrong linking, as ce fi'est point-z-a moi ; ce n'est
pas-t-a lui. Uneducated people, who are familiar with only the sound
of a word, do not always know what consonant, if any, should be
carried forward, and hence they frequently substitute [z] for [t], or vice
versa, or even insert [z] or [t] where no consonant occurs. This is a
pataques, originating in the expression ^e ne sais pas-t-a qu'est-ce. Some-
times the term cuir or velours is used instead, although cuir is confined
rather to the addition of [t] (as in va-t-en ville), and velours is applied
more particularly to the insertion of [z] (as fai eu, pronounced like
Jesus). Examples of such wrong liaisons are very common, as fetais-
t-a la maison, tu es-t-une bete, il etait-z-a Rome, avant-z-hier, fai-z-it6,
moi-z-aussi, il s'en va-t-en guerre, il viendra-z-a Pdques, etc. In the
case of entre quatre yeux, the French Academy has authorized the
pronunciation [atr9katrazj0], more popularly [-katzjo].
^ And occasionally also apres and depuis.
^ Tout is here an indefinite pronoun, but there is no liaison in such
expressions as le tout et la partie, le tout est de savoir, etc., where tout is a
substantive.
176 MANUAL OF FRENCH PRONUNCIATION
je m'en^allais, j'en_ai, on_en_a tant, mais^en disant/
quand^elle chante,^ vraiment^aimable^ tendrement^aime,
tout_-a-fait_extraordinaire, comment_allez-vous ? bientoC
apres^ je ne t'ai point^aime.
But if they do not govern or qualify the following word,
they are not linked to it. Thus, we say Allez-vous^-en, but
Allez-vous I en voiture ? Similarly, there is no linking in such
phrases as the following :
Avez-vous I averti ses amis | ? donnez m'en | un ; va-t-en |
a cheval ; est-on | alle a Rome ; a-t-on | entendu ; eux | ont
ete a Paris.
2. The auxiliary verbs (such as Hre, avoir, devoir, vonloir,
falloir, etc.) are linked to the past participle or infinitive
which immediately follows, as :
II est^arrive, ils_etaient_entres^ il ravait^occupe, d^s qu'il
eut_appris, je devais^aller, il faut_y aller, il fallait^ecrire,
il s'est fait^aimer, 9a peut^etre juste, il veut_y aller.
At the same time, liaisons of this kind do not occur in very
familiar talk, in which it is quite common to hear il est \ arrive,
tu as I eu, nous avons \ eu, tu dois \ ecrire, etc. The use of
the second person singular especially [tutoiement), which in
conversation almost always denotes familiarity or contempt,
does not suit an elevated tone, and consequently does not
accord with frequent liaison. An exception must be made,
however, in the case of tu es, after which liaison is quite
common.
3. The verb is always linked to the pronouns {il, ils,
en, y, etc.), which follow it, as prend^-il, prennent^-ils, ont^-ils
vu, prends^-en, donnes^-en, menes^-y-moi, allez^-y, vas^-y. It is
for this purpose that the imperative singular takes s after e,
when the pronoun en or y is to follow.
4. An adjective, even though polysyllabic, takes liaison
when it precedes the substantive which it qualifies, for in
this case the two words are closely united, making a kind
of compound phrase or indissoluble group ; even two or
1 Liaison is not indispensable in the case of mais, and in mats oui it
never occurs (§ 211 (5)).
2 But when quand is used interrogatively, liaison is not usual, although
permissible, as : Quand etes-vous venu ? [ka st vu vny ?].
WORDS IN COMBINATION 177
three adjectives may be linked provided they are fairly short.
This kind of liaison is particularly useful for marking the
plural. Thus :
Singular : Un grandjiomme, un petit_animal^ un brillant^
orateur, un faux^air, un faux^ebenier, un mauvais^etat,
un pis^aller, un excellent^ecrivain^ un heureux^enfant, le
second^acte, commun^accord, un parfait^honnete homme.
Adjectives in -er also, which have the r mute (especially
premier, dernier), sound it in liaison, the vowel being opened
to [e], as le premier ^aout [I9 pramjer u], le dernier^eleve [I9
dernjsr elsiv], un singulier^effort [de ssgyljsr efoir].^
Plural : De petits_oiseaux^ de beaux^arbres, ses bons^amis,
les jeunes^epoux, certains^auteurs, excellents^amiS; divers^
aspects, plusieursjiommes.
When an adjective follows its substantive, liaison generally
takes place between the two, if the substantive is plural,
Jbut not if it is singular. This method, too, aids in distinguish-
mg the plural. Thus :
Singular : Un avis | important, un portrait | excellent, un vin |
abominable, un nom | ancien, un parfum | agreable, un
officier | estimable, un chaos | affreux.^
Plural : Les langues_etrang^res, contes^espagnols, des jours^
heureux, des rangs^impairs, les gens^ages, Les Etats^Unis,
les aspects^agr cables.^
Note that by such rules an adjective ma}^ frequently be
distinguished from a substantive, as un savant^etranger (a
^ In public speaking and careful speech, the r of verbs in -er is also
sounded as a rule, and the e opened to some extent, as parler^en mon
absence, aimer^oL rive. But this liaison is tending to disappear, and will
soon be confined exclusively to poetry. Even laisser-aller has no liaison.
* Liaison, however, is sometimes made between the singular sub-
stantives sang, rang, aspect, respect, and adjectives which follow, as un
sangjUlustre, un sang^arteriel, un rang^eleve, un aspec{t)_agriable, le
respec{t) Jiumain, but these liaisons are now regarded as pedantic (§ 205).
In aspect and respect, the c is carried forward in the singular only.
' Words ending in mute 5 [cas, repas, etc.), take liaison with z in the
plural, as un cas interessant [ka eteresa], but des cas intdressants
[kaz Eteresa]. Similarly we have un chauffe-pieds excellent [pje sksEla],
but des chauffe-pieds excellents [pjez eksela].
178 MANUAL OF FRENCH PRONUNCIATION
learned foreigner), hut un savant \ etr anger (a foreign scholar).
In the first case savant is the adjective, and etr anger the
substantive ; in the second case it is the opposite. We have
also un sotuaveugle (a foolish blind-man), and un sot \ aveugle
(a blind fool). Similarly, in the plural, we may say un
marchand de draps^nglais, where anglais is an adjective
qualifying draps, but un marchand de draps \ anglais, where
anglais qualifies marchand.
5. Numerals are linked to the words which they
multiply. In this connexion, they are adjectives, and come
under the preceding category. Thus, un^animal, deux^dnes,
trois^enfanis, cinq^amis, sixjieures, sept^oiseaux, huitjiistoires,
neuf^ans, dix^edifices, vingt^etoiles, quatre-vingts^eleves, cent^
ecus, trois cents Jiommes, deux^affreuxjohjets, deux^ et troisjm
[d0ziks e trwazem].^
6. Liaison takes place in certain phrases in fre-
quent use which have practically become compound
expressions, as :
Bout_a bout, chat^echaude, chacun^a chacun, chat^en
poche, croc_-en-jambe, de temps^en temps, de plus_en plus,
de moins^en moins, de mieux_en mieux, de but_en blanc,
de fond^en comble, du haut^en bas, d'un bout^a I'autre,
dos_a dos, deux_^ deux, froid_ai5x pieds, guet_-apens, mot_a
mot, nuit_et jour, pas_a pas, pot_-au-feu, pot_au lait, pot_au
roses, pot_a eau, de pied^en cap, pied^-a-terre,^ petit^a petit,
tout_a coup, tout_a vous, tout_a I'heure, tot^ou tard, vis_-a
vis, etc.
And even accent^aigu, un droit^acquis, etc.
But there is no longer liaison in the following phrases :
Nez a nez, du riz au lait, un chaud et froid, au doigt et a Vceil,
de long en large, un pot a tabac (this last for the sake of euphony).
210. The above six categories contain all the cases where
liaison is considered correct at the present day. But several
other instances of liaison occur whenever the tone of speech
^ The common people even use liaison in the case of quatre by
inserting z between it and a word beginning with a vowel (see § 208,
footnote), as le bal des Quatre Arts [katzair], quatre officiers [katraz
ofisje] .
2 But there is no liaison in such an expression as avoir pied \ a terre,
where the words are not a compound.
WORDS IN COMBINATION 179
becomes more elevated or careful. Thus, the verb and its
complement may be linked, as il est^en ville, cela m'est^egal,
Us ont fait June machine, ditesjun mot, faitesjun beau travail,
and there may be other cases difficult to classify, such as
pret^d sortir, ceux^et elles, des chats^et des chiens, etc.^
211. Note that liaison does not occur :
Xi) In the case of a mute consonant which follows
an r : Thus, de pari en part [da pair d pair], il court encore
^~fil4?u!rak3ir], hordd bord [boir a boir], corps et dme [kor e aim],
le nord-est [I9 norest], le nord-ouest [I9 norwest],^ vers un but
[ver de byt], mort ou vif [moir u vif]. But liaison takes place
with s after r in compound words regarded as single words,
as tiers^etat, or when the s denotes the plural, as leurs^amis,
divers^auteurs, plusieursjiistoires ; with t in verbal forms
in the interrogative, as a quoi ga sertjil ; generally with t in
the adverb fort, as fort^excellent ; and sometimes with s in
toujours, as toujours^est-il gentil. It may also take place
with t in such expressions as mort^aux rats, cela ne sert^d
rien, etc., to avoid a discordant sound.
(2) Generally in the case of the plural s in compound
nouns. In such nouns the pronunciation is therefore the
same in the plural as in the singular, as :
Singular
Un arc-en-ciel
Un char-a-bancs
Un ver-a-soie
Un croc-en-jambe
Un cuiller a cafe
Un guet-apens
Un pot-au-feu
Un pore- epic
Une salle a manger
Un fer a repasser
Plural
des arc(s)-en-ciel
des char(s)-a-bancs
des ver(s)-a-soie
des croc(s)-en-jambe
des cuiller(s) a cafe
des guet(s)-apens
des pot(s)-au-feu
des porc(s)- epics
des salle(s) a manger
des fer(s) a repasser
1 Liaison is usual when one of the conjunctions et or ou unites two
substantives in the plural, the second of which or both of which have
no article, as les ponts^et chaussees, les voiesjet moyens, vertusj&t vices,
vinsjet liqueurs, femmes^ou enfants. But there is no liaison of course in
such an expression as deux heures et demie.
2 Many people, however, carry the d forward in nord-est [nordsst]
and nord-ouest [nordwest], doubtless by analogy with sud-est, sud-ouest.
i8o MANUAL OF FRENCH PRONUNCIATION
(3) Before h * aspirate,' as les hews [le ero], un heros [de ero],
en Hollande [a olaid].
(4) Before a numeral (see § 170), as mes huit oncles [me
qit Slid], les onze animaux [le oiz animo], son onzieme jour
[so ozjem 3uir], Louis onze [Iwi oiz],^ cent un [sa ce], cent
unieme [sa ynjem], quatre-vingt-un [katrgvede]. Notice that
when un is used as a numeral, and not as an article or sub-
stantive, there is no liaison or elision before it : stcr les une
heure [syr le yn oeir], ecrivez le un [ekrive la de].^
(5) Before the [w] of oui and ouate, and the y of many
words when it precedes a vowel [yacht, yatagan, etc.), as mi
oui [de wi], mais oui [me wi], la ouate [la wat], deux yachts [d0
jak], un bon yankee [de bo jaki] ; also before uhlan, as les
uhlans [le yld]. In the case of ouate, however, although the
French Academy recommends that there be no liaison before
it, some well-known writers take the opposite view, and
at all events an e mute is often elided before it, such ex-
pressions as une ouate [yn wat], plein d' ouate [pis dwat],
being common. The word took liaison up to the sixteenth
century at least, and the medical profession speaks of r ouate.
(6) After the conjunction et, as jeune et aimahle [3oen
e Emabl].
212, As already stated, liaison varies according to the
tone of speech. Just as e mute vowels are pronounced more
frequently according as the tone is more elevated, until we
come to verse where they are all pronounced, so the number
of liaisons increases as the tone becomes less familiar, and in
verse as many as possible take place, not only in the interior
of a stress-group, but even between one group and another,
so as to avoid hiatus, which is forbidden there. Thus, in
verse we have les murs pesent^en vain, il faut vengerjun
pere, puisjil sHnterrompit^et dit^d ses disciples. The fact is
that liaison, which is reaUy the reappearance under certain
conditions of a letter not now pronounced, a * dead letter,' is
a result of conservative methods of speech, and this explains
why it is more frequent in reading aloud, in recitations, in
1 But liaison may take place between the t of the verb eire and onze,
as il est onze heures [il et 3:z ce:r], and always takes place in such
expressions as pag{e) onze, pag{e) un, qu ar ant {e)- huit, dix-huit, etc.
2 But if used, e.g. as a substantive, as in tvois un (meaning ' three
ones '), or in cent un (' one hundred ones '), liaison takes place [trwaz&,
satde]
WORDS IN COMBINATION i8i
theatrical oratory, in verse, and generally in elevated speech,
and why the common people manifest a tendency to depart
from it more and more.
213. The fact that in liaison the final consonant of one
word becomes the initial of the next is a frequent source of
confusion to learners. It creates for them, as it were, a
large number of new words, unknown before, and makes it
difficult at first for them to understand the reader or speaker.
For example, they hear such a simple sentence as cette petite
affaire estja moi, but they cannot remember (and for good
reason !) such words as * taffaire ' and * tamoi.' Instances
of the same kind are frequent among French children, who
constantly confuse the limits of words. A proper under-
standing of liaison, with a careful persevering effort to pick
out the true words in spite of the initial added consonant,
will soon enable a beginner to overcome these difficulties.
For the same reason, liaison is an abundant source of puns
and jokes. Thus, it is sometimes asked. Quel est le premier
homme du monde ? Answer, Le rhum de la Jamatque. During
the Revolution in 1790, a humorist denounced a certain
convent as containing cinq canons et vingt-cinq armes. On
examination, the place was found to have cinq dnons et vingt-
cinq carmes I
EXERCISES
(i) Quand_on_est_en_Italie, on parle italien. II me faut^etablir
mes^enfants. Ils_y pensent sans_inquietude. S'il avait_appris
cette nouvelle^ il nous_aurait_aide. Charles a deux^oranges,
j'en^ai cinq. Mes plus_aimables_amis ne sont pas^ici^ ils
sont_en_Espagne. Ces petits^enfants sont^ici sans^abri. En^arri-
vant, nous_avons_entendu le tonnerre. Voici un grosjiomme
avec un grand^ours. Parlez_-en a vos^amis.
(2) The following passage has been divided into stress-
groups and the liaisons marked in the interior of each group
where they occur. The student is recommended to read
it aloud carefully, with due attention to elision and accent
as well as to liaison :
C'etait^a Bologne. | II y avait^eu | une entrevue | entre
le pape | et I'empereur ; | il s'agissait | du duche | de
Florence^ | ou, pour mieux dire, | du sort | de ITtalie. | On_
i82 MANUAL OF FRENCH PRONUNCIATION
avait vu | Paul Trois | et Charles-Quint | causer^ensemble
I sur une terrasse, | et pendant leur entretien | la ville entiere |
se taisait. | Au bout d'une heure | tout^etait decide ; | un
grand bruit d'hommes | et de chevaux | avait succede | au
silence. | On_ignorait | ce qui allait_arriver, | et on s'agi-
tait I pour le savoir ; | mais le plus profond myst^re | avait^ete
ordonne ; | lesjiabitants | regardaient passer | avec curiosite |
et avec terreur | les moindres^officiers | des deux cours ; | on
parlait | d'un demembrement | de I'ltalie, | d'exils, | et de
principautes | nouvelles. | Mon p^re | travaillait | a un grand
tableau^ | et il etait_au haut de I'echelle | qui lui servait^a
peindre, | lorsque des hallebardiers, | leur pique | a la main, |
ouvrirent la porte | et se rang^rent | contre le mur. | Un
page I entra | et cria | a haute voix, | " Cesar ! " | Quelques
minutes^apr^s, | I'empereur | parut, | roide [red] | dans son
pourpoirit, | et souriant | dans sa barbe rousse. | Mon p^re, |
surpris^et charm e | de cette visite | inattendue, | descendait j
aussi vite qu'il pouvait | de son^echelle ; | il etait vieux ; | en
s'appuyant | a la rampe, | il laissa tomber | son pinceau. |
Tout le monde | restait^immobile, | car la presence | de
I'empereur | nous_avait changes | en statues. | Mon p^re |
etait confus | de sa lenteur | et de sa maladresse, | mais_il
craignait, | en se hatant, | de se blesser ; | Charles-Quint | fit
quelques pas_en_avant, | se courba | lentement, | et ramassa |
le pinceau. | " Le Titien, | dit_-il, | d'une voix claire | et
imperieuse, | le Titien | merite bien | d'etre servi | par Cesar."
I Et avec une majeste | vraiment | sans^egale, | il rendit | le
pinceau | a mon p^re, | qui mit_un genou en terre | pour le
recevoir. | A. de Musset, Le Fils du Titien
(3) The student is invited to divide the following passages
into stress-groups for himself and note the liaisons. Each
passage, after being thus treated, should be read aloud
carefully.
(i) II n'y a rien au monde qui se fasse tant admirer qu'un
homme qui sait etre malheureux avec courage. — Seneque.
(2) Les peuples jeunes ne voient que ce qu'ils ont a gagner, les
vieilles nations songent a ce qu'elles ont a perdre.
(3) La faveur qu'on merite est toujours achetee. — Corneille.
(4) La delicatesse est pour les ames elevees un devoir plus
imperieux que la justice. — Mme de Stael.
WORDS IN COMBINATION 183
(5) II est plus honteux de se defier de ses amis que d'en etre
trompe. — La Rochefoucauld.
(6) II est plus aise de juger quelqu'un apr^s une heure de
conversation dans un salon qu'apr^s dix ans de vie commune.
(7) La politesse, c'est Part de faire ce qui vous ennuie comma
si cela vous amusait.
(8) Quand on rit d'un obstacle, il est presque vaincu. — ■
C. BONJOUR.
(9) Toutes les passions sont exagerees ; elles ne sont des
passions que parce qu'elles exagerent. — Chamfort.
10) Affecter de savoir ce qu'on ignore, c'est tendre un pi^ge
dans lequel le plus leger incident pent vous faire tomber.
11) Si vous voulez etre riche, n'apprenez pas seulement
comment on gagne, sachez aussi comment on menage.
— Franklin.
12) Aux yeux des partis, qui cesse d'etre un esclave devient
un deserteur. — J. Simon.
13) C'est jouir du bonheur que de voir sans envie le bonheur
des autres et avec satisfaction le bonheur commun. —
BOSSUET.
14) II y,a de violents outrages que Ton oublie, et des paroles
maladroites que Ton ne pardonne jamais.
15) L'ame qui n'a pas de but etabli s'egare et se perd ; c'est
n'etre en aucun lieu que d'etre par tout. — ^Montaigne.
16) La dignite de notre espece n'est pas moins attestee
par les oeuvres du coeur que par celles du genie.
17) Les passions, qui sont de bons auxiliaires, sont de mauvais
conseillers. — Balmes.
18) II est impossible de porter le flambeau de la verite dans une
foule sans bruler la barbe a quelqu'un.
CHAPTER XXXIV
EMPHATIC OR SUPPLEMENTARY ACCENT
(accent d'insistance)
214. Emphasis is the special stress given to any part
of a sentence to which the speaker wishes to call particular
attention. It is of every degree, from the most forceful to
where it ceases to be emphasis. Its place in a sentence is
not fixed, as that of the rhythmic accent is. It falls wherever
the meaning requires it, occasionally on phrases or on state-
ments of some length, but generally on single words. It
throws the syllable on which it falls, and consequently the
word, into greater prominence than what precedes or follows,
and in this way sometimes changes the meaning considerably.
Thus, the English sentence, " Did you motor there yesterday ? "
may have five different meanings according as the emphasis
is placed on one or the other of the five words. " Did you
motor there yesterday ? " has quite a different meaning from
" Did you motor there yesterday ? " It is this emphasis on
individual words that we deal with in this chapter. It is
extremely frequent in ordinary French conversation or
oratory, because the French tend to be emotional, animated,
or emphatic in their speech ; and the student will thus
see the necessity of practising it carefully and perseveringly,
until he acquires facility in it, as no French pronunciation
can be really good without it.
215. How is such emphatic accent expressed in French ?
It depends upon circumstances. There are different cases to
consider :
I. When the Word has more than one Syllable
In such cases the emphasis is scarcely ever expressed by
reinforcing the rhythmic accent, if the word has one, as an
English student would imagine should be done. This would
184
WORDS IN COMBINATION 185
spoil the language by robbing it of its variety. Occasionally
it may be expressed by stressing every syllable in the
word, as par-faite-ment ! But generally, indeed in almost
every case, it is done by putting more stress on another
syllable than the last one, and thus bringing the word
into special prominence. For example, in the sentence,
Cest un miserable, cet homme, uttered normally and without
emotion, there is the usual slight stress on the last syllable
of miserable, but as soon as the word is pronounced with
emphasis or feeling, as frequently happens, a separate stress,
much stronger than the other, falls on another syllable
(in this case, the first one), as c'est un miserable, cet homme!
The rhythmic or tonic stress still remains on the last syllable,
but the emphatic stress is more intense, and therefore more
noticeable. There are many words which, owing to their
very nature, are more frequently spoken in this way than
normally. This is so with emotional words, expressing
fear, joy, grief, astonishment, etc. Among these are ad-
jectives like terrible, effrayant, incroyable, desolant, malheureux ;
adverbs like beaucoup, extremement, parfaitement, certaine-
ment ; verbs like fremir, pleurer, hurler ; nouns like scelerat,
miserable, assassin ; and practically all words of abuse, such
as cochon, fripon, gaffeuse, etc.
216. The question as to which of the syllables, other than
the last one, should take the emphatic accent, depends on
certain conditions, as follows :
(i) If the word forms part of a rhythmic element, i.e.
if it does not stand alone, isolated as it were, the syllable
stressed is usually the first one commencing with a
consonant, as : c'est repugnant, je suis si malheureux, c'est
absolument vrai. The reason why the first syllable, if it
begins with a vowel, is passed over, is that in this case it
would be necessary to include in the stress the final consonant
of the preceding word, and this would sound rather absurd
and even amusing. Further examples are :
C'est epouvantable, quelle barbaric, c'est un imbecile, c'est
un vol magnifique, c'est impossible, il est assommant, il s'en-
fongait dans son lit, vous etes un assassin.
In such a case the initial consonant is made very long, being
generally more than double in length. Thus barbaric becomes
i86 MANUAL OF FRENCH PRONUNCIATION
[biarbari]. While the rhythmic accent does not permit of
any consonantal lengthening, the emphatic accent finds
its peculiar characteristic in the fact that the consonant is
vigorously seized upon and prolonged (hence the French
name, accent d'insi stance), and this should be remembered by
students, as English emphasis does not generally put stress
on the consonant.
In addition, the vowel of the emphasized syllable is in-
creased in height and intensity ; but it is not increased in
duration, i.e. lengthened, except in cases where it is already
long or half-long (see 104 (3), (4) 3). Thus, in the expression
il y en a heaucoup, the b is very sensibly prolonged ; and
the vowel following, being already half-long, takes on full
length or even more [boiku], as it does likewise in il pleurait,
il est vivant, etc. If the vowel be short, it cannot be
lengthened, although the other changes referred to occur,
as pa,rfaitement}
217. (2) If the word forms a rhythmic element by
itself, i.e. if it is isolated, it takes the Emphatic Accent on
the first syllable, whatever it be, as :
Cat homme est fou. — Absolu-
ment.
Je crois qu'il mentit. — Abomi-
nablement.
II est tr^s gentil. — Extreme-
ment.
Attention J s'il vous plait.
Aujourd'hui^ je me moque de
ces choses.
Compare the following examples, which show clearly the
difference between such cases and the previous ones :
C'est incroyable. — Incroyable, mon cher monsieur, incroy-
able.
Ce discours est assommant. — Assommant.
Le bruit etait epouvantable. — ^pouvantable.
The emphatic stress must not be placed on any other
syllable, except under exceptional circumstances. The forms
epouvantable and epouvantable are both correct, each in its
^ An exception is [e], which, may take on extra length (§ 20).
Cette femme est stupide. — ^Com-
pl^tement.
Voulez-vous le voir ? — Certaine-
ment.
Mais il I'a dit. — Precis ement.
C'est un immense obstacle. —
Formidable !
Misericorde 1
WORDS IN COMBINATION 187
place, but epouvantable would not be so under normal
circumstances.
Where the initial syllable commences with a vowel, the
consonantal lengthening, which is such a feature of the
emphatic accent, is not wanting. In such cases the consonant
or group of consonants following the vowel is prolonged,
although not quite so much as when a consonant commences
the word.
II. When the Word is Monosyllabic
218. In such cases, if the word has no rhythmic accent, the
emphasis is of the same nature as that already described,
consisting in the lengthening of the consonant, and an in-
crease of height, intensity, and (in allowable cases) of duration
in the vowel. But if the word has the rhythmic accent, there
is no increase as a rule in the height or intensity of the vowel.
Sometimes the word is quite small and indeed insignifi-
cant apart from the emphasis, as :
C'est la meme personne, ce n'est pas vrai^ ce n'est rien,
voila trois jours que nous ne parlons que d(e) cette affaire.
In this last example, the word trots would otherwise be
unaccented, being an adjective of number, and coming more-
over before an accented monosyllable (§ 202), but it here
receives the emphatic accent because the speaker wishes
to draw attention to the time occupied. The word que, too,
is not only an unaccented word naturally, but has an e mute
vowel which might be elided in ordinary circumstances, as
nous n{e) parlons qu{e) de cette affaire ; but here the e mute
reasserts itself and the word is forcibly stressed in order to
make prominent the fact that a certain affair has been talked
of to the exclusion of all others. Further examples are :
Vous etes le sel de la terre ; il faut faire juste ; regardez
bien ; vous, maintenant ; c'est bien fait ; il n'y a que deux
sortes de geas au monde ; c'est si amusant ; voila tout ce
que je te permets. L'Ecosse est le plus beau pays que j'aie
vu. Fais ton pain^ je n(e) te nourris plus. Vous I'avez dit ?
Non, je ne I'ai pas dit.
219. We have stated that the emphatic accent does not
destroy the rhythmic or tonic one on the last syllable of a
i88 MANUAL OF FRENCH PRONUNCIATION
stress-group : it is merely supplementary to it. Thus in
c'est epouvantable, cette affaire, uttered with emotion, there
are two accents on epouvantable — the emphatic on -pou-, and
the usual one on -table. Similarly in je reste, tu fen vas,
je and tu have the emphatic accent, but this does not prevent
the usual one on reste and vas. The two accents each serve a
different purpose. The rhythmic one is to be expected : it
is normal and regular, and can produce no effect beyond
its rhythmic function. The emphatic accent, on the other
hand, is unexpected : it comes forcibly, sometimes indeed
violently, and serves to draw immediate attention to the
word on which it falls. It acts so suddenly and strongly that
one would sometimes fancy that the rhythmic accent has
disappeared. But the latter is still there, and only under
certain conditions of weakness is it diminished. One of
these conditions is the end of a sentence. If the em-
phatic syllable be near the end — say the last but two — the
tonic stress at the end is apt to suffer. This is due to the
habit of letting the voice drop, combined with the special
energy just expended on the emphasis. But the diminution
of stress is only slight, for the speaker's object is to make
the word prominent, and he consequently does his best to
sustain all its syllables. Again, in those cases where the
emphatic stress faUs on a syllable immediately preceding
the tonic stress, there is usually some diminution of the latter,
arising from the difficulty that always exists in making an
effort twice immediately in succession. If the word happens
also to be the last one in the sentence, the diminution is
sometimes considerable, although never so great as to cause
the disappearance of the rhythmic or tonic accent. The
final syllable always remains clear and distinct, and more
intense than an unaccented one.
220. Another method of emphasis remains to be referred
to. Sometimes words or groups of words occur in series.
They may for instance form an enumeration, or a gradation,
or be placed in opposition to each other. In such a case
the emphatic accent, instead of being placed on the really
important words, may be placed on the initial syllable,
whatever it may be, of the various groups which corre-
spond to each other, and consequently sometimes on a word
quite insignificant in itself. The effect produced is the same,
WORDS IN COMBINATION 189
for the attention is immediately drawn to the whole group by
emphasizing an insignificant word at the commencement
of it. Sometimes either method may be chosen. We give
examples which will easily be understood without further
explanation :
Enumeration
J'ai etudie le vegetal dans tons ses mysteres, dans la tige,
dans le bourgeon^ dans le sepale, dans le petale, dans
I'etamine, dans le carpelle, dans Tovule, dans la th^que,
dans la sporange^ et dans I'apothecion. — Victor Hugo,
VHomme qui rit.
J'entre dans un cabinet de lecture et je lis a haute voix en
parcourant les casiers. Revue commerciale, Revue litteraire,
Revue des Cours, Revue historique, Journal des Savants,
Revue des Deux Mondes, Bulletin financier, Revue arche-
ologique. Journal des Rentiers, Bulletin hebdomadaire,
Journal de la Marine.
Gradation
II avait remarque la superposition des fleaux, les rois sur
le peuple, la guerre sur les rois, la peste sur la guerre, la
famine sur la peste, la betise sur le tout. (Or one might here
say, '' les rois sur le peuple, la guerre sur les rois, la peste
sur la guerre, la famine sur la peste, la betise sur tout.")
C'est dans la politique que se font, defont, et surfont les
reputations.
Allez-vous au theatre ce soir ? Impossible, monsieur,
tout a fait impossible.
Opposition
Donnez et pardonnez. L'homme propose, et Dieu dispose.
II ne donnait de satisfecit a qui que ce soit ni a quoi que
ce soit.
Ce que vous appelez arranger, je I'appelle deranger.
Dechirez vos cceurs et non vos vetements.
Faut-il dire et redire la meme chose ?
II y a beaucoup de difference entre les conservateurs et
les^opposants.
190 MANUAL OF FRENCH PRONUNCIATION
From the last example it will be noticed that when once
the emphatic stress has fallen on the first syllable of one of
the words in a series, it falls on the first syllable of all those
which follow, even though beginning with a vowel, and
in this case any consonant carried over in liaison takes
its part in the emphasis and provides the consonantal
lengthening.
EXERCISE
The following extract will afford abundant illustrations
of the rules in this chapter. Obviously, the emphasis might
be differently placed according to the individual peculiarities
of the speaker, but apart from such variations the stresses
represent the pronunciation of a careful reader or speaker.
Le plus souvent, | je partais | pour le college | a jeun, |
I'estomac | et la tete | vides. | Quand ma grand'm^re | venait
nous voir, | c'etait le bon jour ; | elle m'enrichissait | de
quelque petite monnaie. | Je calculais | alors | sur la route |
ce que je pourrais bien acheter | pour tromper | ma faim. |
Le plus sage | eut ete | d'entrer | chez le boulanger ; | mais
comment | trahir | mapauvrete | en mangeant | mon pain sec |
devant mes camarades ? | D'avance, | je me voyais | expose |
a leur rire, | et j'en fremissais. | Cet age | est sans pitie. |
Aujourd'hui, | cette indigence | nee | de la persecution, |
fierement, | noblement | supportee | par les miens, | fait ma
gloire. I Alors, | elle me semblait | une honte, | et je la
cachais | de mon mieux. | Terrible respect | humain. |
Pour echapper | aux railleries, | j'imaginais | d'acheter |
quelque chose | d'assez substantiel | pour me soutenir, | et
qui ressemblat | pourtant | a une friandise. | Le plus sou-
vent, I c'etait le pain d'epice | qui faisait les frais | de mon
dejeuner. | II ne manquait pas | de boutiques | en ce genre |
sur mon chemin. | Pour deux sous | on avait un morceau |
magnifique, | un homme superbe, | un geant | par la hau-
teur I de la taille ; | en revanche, | il etait si plat, | que je
le glissais | dans mon carton, | et il ne le gonfiait gu^re. |
Pendant la classe, | quand je sentais | le vercige | me saisir, |
et que mes yeux | voyaient trouble | par I'effet | de 1' inani-
tion, I je lui cassais | un bras, | une jambe, | que je grigno-
tais I a la derobee. | Mes voisins | ne tardaient gu^re | a
surprendre | mon petit manage. | " Que manges-tu la ? " |
WORDS IN COMBINATION 191
me disaient-ils. | Je repondais, | non sans rougir, | " Mon
dessert." |
On dit I que les souffrances | physiques | sont bonnes | a
Tame. | On voit | que j'etais dans I'etat | le plus propre | a
developper | la mienne. | Mais le corps, I lui, | a ete de-
prime. I Malgre | les adoucissements | qui sont venus | plus
tard, I je porte | toujours | ce temps | en moi. | . . . Mes priva-
tions I peuvent se resumer | en trois mots : | jusqu'a quinze
ans, I point de viande, | point de vin, | point de feu. | Du
pain, I des legumes | le plus souvent | cuits a I'eau | et au sel. |
Si j'aisurvecu, | c'est que malgre | les souffrances | etla sante j
ruinee | de ma m^re, | la saine constitution | de mon p^re |
prevalut | en moi. | Le travail, | les habitudes | de la vie |
solitaire | que je menais | avec mes parents, | me soutinrent |
aussi, I me rendirent | actif, | mais sans me fortifier | jamais, |
de sorte | que ma chetive figure | reste | comme un monu-
ment I de ces temps de deuil. | Les cicatrices | que garde |
ma main droite | temoignent | des temps d'hiver | passes sans
feu. I Et cependant, | parmi les coups | et contre-coups |
qu'un enfant | semblait | ne pas pouvoir | supporter, | je
restai | pour les voir, | et vivant | pour les raconter." |
Jules Michelet
CHAPTER XXXV
ASSIMILATION
221. Frequently, when two sounds (either vowels or con-
sonants) come together, one of them tends to become as-
similated to the other, borrowing part of its characteristics,
so as to avoid a sudden change in the position of the vocal
organs. Vocalic assimilation has already been referred to
(§ 29). We have it in such words as etais, eleve, ehene, where
the first vowel (e) is often pronounced half -open or open [e],
owing to the tongue anticipating the position of the second
vowel by descending. But further reference need not be
made to such cases, as assimilation between vowels is not
nearly so frequent as between consonants. Instances of
assimilation between consonants are common both in English
and French. Thus, in English, the plural s, while it retains
its proper sound in such words as ropes, docks, butts, practically
becomes [z] in robes, dogs, buds, i.e. it is voiced through
assimilation to the voiced consonant preceding. In the same
way, in French, two adjoining consonants frequently have
an active influence on each other. The characteristics of
the one pass in part or even completely to the other, with
the result that considerable modifications of sound may be
produced. In some cases where the assimilative process
is very old, it shows itself in the ordinary spelling, as in
chercher, which has taken the place of the ancient form
cerchier. But there are numerous cases where the assimila-
tive influence exists without any change in the spelling.
Thus, we have anecdote [anegdot], Strasbourg [strazbuir],
transvaser [trazvaze], in all of which the first of the two
consonants is voiced to make it agree with the second ; and
abces [apse], medecin [metsg], savetier [saftje], chemin de fer
[/omEtfeir], in all of which the first is unvoiced for the same
reason.
222. It is this kind of consonantal assimilation to which
192
WORDS IN COMBINATION 193
we refer in this chapter. It is common in French, especially
in rapid speaking, where economy of effort is needed. It
only occurs, however, when the two consonants come into
immediate contact ; and an e mute between them does
not of course prevent this, provided the e mute be elided in
pronunciation.^ Moreover, the assimilation consists as a
rule in the first consonant being influenced by the second,
as in the examples just mentioned (compare English cwp-
board). Such assimilation is generally called regressive,
but the term ' anticipative ' seems preferable, for the modi-
fication in the first consonant is not exactly due to some
influence from behind, but rather to an anticipation, conscious
or sub-conscious, of the second consonant, an act of foresight,
as it were, on the part of the organs of perception, which put
themselves in position well in advance. When, as occasion-
ally happens, the second consonant is influenced by the
first (compare English dogs, observe, etc.), the assimilation is
known as progressive, but this kind only occurs in French
as a rule when the second consonant is /, m, r, or one of the
semi-consonants [w, j, q], as clef [kje], peuple [poepj], quatre
[katj-], pied [pje]. In such cases the second consonant
loses voice under the influence of the first (see § 186 (2)).
This kind of assimilation, however, need occasion no difiiculty.
For all practical purposes it may be neglected, as the voice
will usually disappear unconsciously.
223. The regressive cases of assimilation may be divided
into two classes :
I. In the interior of words or well-known phrases :
Here the assimilation is often of a complete kind, i.e. not
only is there a voicing or unvoicing of the first consonant
to make it correspond with the second, but the mode of
assimilation, the force of utterance, etc., become assimilated
also. In other words, the first consonant becomes practically
of the same type as the second. Thus, in chemin de fer
[fametfeir], as popularly pronounced, the d not only loses
its voice, but really becomes a t. It should be remembered
that there is a difference : an unvoiced d is not exactly a t,
for the former has the point of articulation slightly different
and is also pronounced with much less force ; nor is a voiced
^ Thus, in second, if the e be ehded, we have [zgo], but otherwise
[sago].
N
194 MANUAL OF FRENCH PRONUNCIATION
/ practically the same as a d, for the same reasons.^ But,
as just stated, when assimilation occurs in the interior of a
word or group of words closely connected, it is not always
restricted to the mere voicing or unvoicing of the first con-
sonant (although this is the most frequent mode of assimila-
tion in other cases), but is often of a more complete character.
This is particularly the case in rapid conversation, where
direct or sudden changes in the organs of speech are frequently
neglected. In slow or elevated speech, on the other hand,
complete assimilation would shock the ear, and would be
condemned by all masters of diction ; and this should be
remembered by the student in his desire to acquire a good
pronunciation. Examples :
Gibeci^re [3ipsJ8ir]
Jeter [/te]
Sauvetage [softai3]
Clavecin [klafse]
Gisivete [wazifte]
Rejeton [rajto]
Eczema [egzema]
Second [zgo]
Afghan [avga]
Susdit [syzdi]
De temps en temps [tiazata]
Rez-de-chaussee [ret/ose]
Valet de chambre [valet/dibr]
Chauve-souris [fofsuri]
Garde champ etre [gartJapEitr]
Haut de forme [otform]
Haut-de-chausses [ot/ois]
Au-dessus, la-dessus [otsy, latsy]
Coup de pied [kutpje]
Sauve-toi [softwa]
224. Note the following special cases in the interior of
words :
(i) The prefixes ab-, ob-, sub- : Here the h practically
becomes p before a voiceless consonant, as abstenir [apstgniir],
obtenir [optoniir], subterfuge [sypt8rfyi3]. Other examples:
absent, absinthe, absolu, abstrait, observer, obscur, obseder,
substance, subtil, substituer, etc. The word subsister, how-
ever, is frequently pronounced [sybziste], with progressive
assimilation.
Similarly, in the prefixes dis-, trans-, the s becomes
[z] before a voiced consonant, as disgrace [dizgrais], disjoindre
[diz3W8idr], transgresser [trozgrese], transborder [trazborde].
Other examples : disgregation, disjonctif, transvaser.
(2) When s precedes b or m : Here the assimilation may
1 In phonetic transcript, mere unvoicing may be denoted by o under
or over the consonant, as [^], [b], etc., and mere voicing by v, as [t],
[p], etc.
WORDS IN COMBINATION 195
either be regressive or progressive, but the regressive is the
more usual with b, and the progressive with m, as sbire
[zbiir], presbytere [prezbiteir], asthmatique [asipatik], en-
thousiasme [atuzjasiji].^ Other examples : asbeste, Lisbonne,
Ratisbonne, Strasbourg, jasmin, prisma, lyrisme, rhumatisme,
cataplasme, mutisme, hero'isme, Christianisme, etc. When
thus unvoiced, the final m is so faint that it frequently dis-
appears in popular speech (compare r, I, § 186 (3)).^ This
is specially so in long words, where the [sm] becomes merely
a more or less prolonged [s]. Hence a common word
like rhumatisme has three pronunciations, [rymatism], or
[rymatizm], which are equally good, and [rymatis], which is
only used by * the people.'
When sm is initial (which occurs only in foreign words),
it is always pronounced [sm], as in smalah, smilax, Smyrne,
etc.
(3) When m is initial : Here, as in the preceding cases,
it frequently loses its normal voiced form, and becomes [m].
The word monsieur, e.g., is usually pronounced [m9sj0], with
the m voiced, but frequently in hurried speech the e mute is
elided, and the m, coming under the influence of the s, loses
its voice, the word becoming [msj0]. In such cases the
unvoiced m in careless speech sometimes loses its nasal sound
and becomes a p, as [psj0].
(4) When s follows 1 : Here there is a progressive as-
similation, the s being voiced, as Alsace [alzas], alsatique
[alzatik], balsamier [balzamje], balsamine [balzamin], Belsunce
[bdzdeis].
225. 2. Between words : In this case the assimilation is
only partial, not complete. It is restricted to the mere
voicing or unvoicing of the first consonant, as une grande
tasse [yn grattais], une pauvre femme [yn poffam], une robe
sombre [yn ropsoibr]. In such cases the first consonant does
not exactly become the same as the second, although it may
be written so phonetically. In such a sentence as je viens
^ It is in Switzerland and Belgium that the regressive assimilation is
mostly found with m [azmatik], etc,
2 The same phenomenon appears, in popular speech, in words in
-iste {artiste, anarchiste, etc.), which tend to be pronounced as if the
final was -isse [artis], [anarjis]. Even such words as prdtexte, insecte,
etc., become [pretsks], [essk], etc. This is a freedom of language,
however, that should be avoided.
196 MANUAL OF FRENCH PRONUNCIATION
d{e) servir, the d does not become a t, but only a more or less
unvoiced d [d] ; otherwise it would be identical with je viens
t(e) servir. Any speaker who would make the assimilation
between words more than partial would utterly spoil both
his meaning and pronunciation, and would be put down at
once as a foreigner. Examples :
Consonants of same order
II frappe bien [frabbje]
Un esclave fugitive [£sklaffy3itif]
Une tete d'homme [tsddom]
Tu faches Georges [fcL33or3]
II navigue constamment [navik-
kostama]
Ne tombe pas [no toppa]
Une route droite [ruddrwat]
Une pipe blanche [pibblaij]
Une vache jaune [va33oin]
Une page charmante [paj/armdit]
Un vase sacre [vassakre]
Un singe superbe [sejsyperb]
Consonants of different order
Je te crois [Jtakrwa]
^feglise protestante [eglisprotes-
tait]
Grog chaud [grok/o]
Un brave coeur [brafkoeir]
lis savent tout [saftu]
Chaque jour [Jag3uir]
Est-ce bien [ezbjs]
Un bee d'aigle [bsgdegl]
L'archeveque de Paris [lar/aveg-
dapari]
Quinze francs [kesfrci]
Cap Vert [kabveir]
Et ainsi de suite [eesit sqit]
Care must be taken, as we have said, to make the assimila-
tion only partial, consisting of the mere voicing or unvoicing.
The difference thus produced is evident by comparing such
expressions as tete carree, tete dure ; un Arahe deguenille,
un Arahe cruel ; il se repose beaucoup, il se repose toujours,
etc.
226. The above two classes do not entirely exhaust the
subject. In addition to voicing or unvoicing, we sometimes
have sounds becoming nasalized or denasalized through
assimilation. Any vowel or consonant, for example, placed
between nasals, tends to become more or less nasalized itself,
as moi-meme [mwameim], maman [mdma], rongeant [r53a],
en venant [avna]. The Americans in particular are inclined
to nasalize all vowels in contact with nasal consonants, as
[5m], [yn], [kan], etc. Sometimes, when a plosive is nasalized
in this w^ay, it is turned into its corresponding consonant
(see § 90) . Thus, d becomes n, h becomes m, and g becomes
the English [g], as pendant [pana], lendemain [lanms], une
tombe neuve [tomnoeiv], une longue main [logme]. The reason
WORDS IN COMBINATION 197
for this is that there is no difference, for example, between
d and n except the lowering of the velum for the latter, and
as it has to be lowered for the vowel preceding and the one
following, it is only natural that it should tend to remain
down for the consonant between them. At the same time,
such influences as those we refer to in this paragraph should
be resisted, even in rapid or familiar speech. They are
regarded as provincial or dialectical, and the form of words
produced is less harmonious than the correct one.
Finally, all tendency to assimilation should be resisted
beyond what is really natural and well recognized. The
changes referred to are spontaneous and involuntary. They
are only legitimate when the speaker is unconscious of them.
If a learner is rather halting or embarrassed in his French,
as is very probable, the conscious use of assimilative forms
will only make it disagreeable or even grotesque to polite
ears.
CHAPTER XXXVI
INTONATION
227. By Intonation is meant the rising and falling of the
voice, in other words, its musical movement. In solemn
reading there is usually not much of this : the voice merely
rises to denote an interrogation or exclamation, and falls
at the end of sentences. But in ordinary speaking the voice
rarely maintains itself on the same musical note : it is con-
stantly moving upward and downward through a certain
number of notes. This is the case more or less in all
languages, every word-group having a melody of its own.
It follows that if the intonation is not good, the utterance
is unmelodious and unattractive. It is almost impossible
to lay down specific laws in the matter, as the risings and
fallings are sometimes very delicate. All that we propose
to do in this chapter is to give some general indications of
the movements in simple declarative sentences, excluding
interrogative and exclamative ones, which are dealt with later.
228. Every declarative sentence consists of two parts,
the first having an upward inflexion, denoting incompleteness
of statement, and the second a downward one, denoting
completeness. This is specially noticeable if we take simple
sentences containing only two stress-groups, such as the
following :
Si vous voulez, || je viendrai.
Cette table || est lourde.
On sonnait le couvre-feu.
Je I'ai entendu, || mais je ne le
crois pas.
Je me promenais || dans le jardin.
II demande || de I'argent.
In each of these the voice rises to the end of the first group,
suggesting that the statement is still in suspense, and then
falls to the end of the second, implying that the statement
is now being concluded. The movement in such simple
198
WORDS IN COMBINATION 199
cases might be roughly expressed by two obHque Hnes,
thus : '
Usually, however, sentences are longer than these, and
each of the two parts may contain any number of stress-
groups. Sometimes each part has an equal number. In
Alexandrine verse, for example, when a line embodies a
complete statement in itself, the hemistichs contain two
stress-groups each, the first hemistich taking the upward
inflexion and the second the downward one, as :
Son ombre | vers mon lit || a paru | se baisser.
In prose too there is frequently an equality of groups, and
occasionally this may mount up to a dozen or more groups
in each part of a sentence. Bossuet's writings contain a
large number of such, as :
A mesure | qu'il approchait, || je le voyais | disparaitre.
II n'y a point | de puissance | humaine || qui ne serve | a
d'autres desseins | que las siens.
Celui I qui insultait | a I'aveuglement | des autres || tombe |
lui-meme | dans des ten^bres | plus epaisses.
Generally, however, each sentence has an inequality of
groups, as in the following examples, where the last group
alone forms the downward part :
Tout bonheur que la main n'atteint pas |1 est un reve.
II n'y a rien que les hommes aiment mieux conserver et
qu'ils menagent moins || que leur propre vie.
229. As stated above, in all such cases of simple declaration,
the first part of the sentence, viz. that which leaves the
meaning in suspense, is spoken with the upward inflexion so
as to sustain the attention, while the second part, which
satisfies or removes the suspense, is uttered with the down-
ward inflexion, which makes the hearer feel that the sentence
is being concluded. If we intended to say // n'y a pas \\ de
bonheur, the expression il n'y a pas would of itself leave the
200 MANUAL OF FRENCH PRONUNCIATION
meaning incomplete and in suspense : it must therefore take
the upward inflexion. The expression de bonheur adds the
completeness intended and removes the suspense : it must
therefore take the downward inflexion. If we wished to
lengthen the statement and say // n'y a pas de bonheur \\ au
sein de Vinsincerite, the expression il n'y a pas de bonheur
would be incomplete, still leaving the meaning in suspense :
it must therefore take the upward inflexion, while the re-
mainder of the sentence takes the downward one. If we
wished to say still further, // n'y a pas de bonheur au sein de
Vinsincerite, \\ parce qu'on n'y trouve pas de repos, the whole
sentence up to the final syllable of insincerite now takes the
upward inflexion for the same reason, and the rest takes the
downward one. Similarly, we might have the following
sentences, each complete in itself :
La joie || se trouve.
La joie se trouve || au fond de toutes choses.
La joie se trouve au fond de toutes choses, || mais il appartient
k chacun de Fen extraire.
230. Such examples as we have given are sufficient to
illustrate the intonation or musical movement of ordinary
declarative sentences. The punctuation marks of common
orthography correspond to some extent to this movement.
Generally speaking, a comma or semi-colon denotes an upward
inflexion, and a full stop indicates a downward one, while
the clauses between commas take a dependent place in the
general movement. Such rules, however, are by no means
absolute, as punctuation marks are meant for other purposes.
In long sentences, the movements can never be fixed and
determinate, as speakers differ from each other in sub-
ordinate clauses, and it would be absurd to prescribe definite
rules in such a case. But with all good speakers the main
characteristics of the movement remain, however long or
involved the sentence is. What these characteristics are
will be still more evident from a separate examination of each
of the parts :
231. I. The Upward Part: In this part the first
stress-group rises gradually to a certain height, sometimes
amounting to four or five notes. The second group, and the
following ones, whatever their nature, maintain this height
WORDS IN COMBINATION
201
generally. They each commence on a somewhat lower note
than the previous group ends on, but their tonic syllable rises
generally to the highest note of that group, although not
usually exceeding it. The last group commences like the
others, but it always finishes on a rising note as high as any
preceding ones, and in many cases even higher. Conse-
quently, if we take such a sentence as :
On y voit | dans ces prairies | errer a I'aventure || des trou-
peaux I de trois ou quatre mille | buffles sauvages,
the whole upward part has a form something like the
following :
Fig. 7
The special characteristic of this part is that throughout
it, from the first syllable to the last, there are no sudden
movements or leaps, but a well-marked undulation, without
anything resembling angles. Whatever alteration is made
takes place by degrees or glides. In this respect speaking
differs from singing, in which the notes generally follow each
other by leaps. In speaking, the voice glides through the
intervening notes : it does not pass from a higher note to a
lower one by a sudden drop, but meets it half-way, as it
were, and the movements can only be . represented as above
by a curved line, which in music would require the term
glissando. No doubt there is less of the glide in French than
in English, but it is a characteristic to some extent of all
speech, particularly of the upward intonation.
232. 2. The Downward Part : This part commences with
a sudden drop, without any intervening notes or undulation.
Beginning with its first syllable, therefore, the whole of it
is generally spoken on a lower level. At the same time all
the groups contained in it, except the last, take the same
movements as those in the upward part, i.e. the first group
gradually rises a few tones and the rest follow with frequent
undulations. But the general characteristic is that the
low notes sink lower than those in the upward part, and the
202 MANUAL OF FRENCH PRONUNCIATION
high notes do not rise so high (see Fig. 8 below). There is
thus a gradual fall from group to group, the fall amounting
to one note at least in each group. The last group reaches
its end frequently by a gradual descent rather than an un-
dulation, and finishes on a much lower note than the upward
part commenced on. The fall from the highest note of the
whole sentence to the concluding one is generally six or
seven notes, but may sometimes be more than an octave.
In the case of many excellent French speakers, indeed, the
last note often sinks so much that it practically loses its
sonority, and becomes breathed or whispered. This is
particularly the case with the high vowels, i, y, u (§ 102),
and sometimes happens even in short expressions {il est parti,
il y en a beaucoup, etc.). The whole sentence, including both
parts, has thus a movement resembling the following :
Fig. 8
Though it may sometimes happen that several successive
sentences in a paragraph are declarative ones, terminating
with the downward inflexion, yet this inflexion need not
become monotonous ; because, as every sentence may vary
in the commencing note, it may likewise vary in the con-
cluding one. Due attention to this will prevent the re-
currence of any wearisome sameness of tone at the end of
every sentence.
233. A parenthetical or incidental clause does not affect
the construction of the sentence within which it is inserted,
and consequently only takes a subordinate part in the musical
movement referred to. It requires a pause before and after
it, so as to give it an isolated and independent character,
and is pronounced with a depression of voice and somewhat
faster than the rest of the sentence ; but it generally rises
a little towards the end, in conformity with the general
intonation. Examples :
Je I'avais attrape, continua le bandit, et cela me faisait
plaisir.
WORDS IN COMBINATION 203
II lui donnera^ repondirent les ambassadeurs, sa fille et la
moitie de son royaume.
A ces mots ranimal pervers
{Cest le serpent que je veux dire
Et non V Homme ; on pourrait aisSment ^'y iromper),
A ces mots, etc.
234. The taking of breath does not in any way modify
the intonation. Breath should be taken, as a rule, at the
end of every sentence. If the sentence is rather long, it
may be taken between the two parts, or if necessary between
any two stress-groups ; but on no account should it be taken
in the middle of a stress-group. The breathing should not be
clavicular, but rather by depression of the diaphragm ; and
the student should practise the power of holding the breath,
and expending it slowly and equally as the sentence is uttered.
EXERCISES
I. Read the following sentences aloud with the proper
intonation :
(i) L' eloquence du coeur || persuade aisement.
(2) La cause du faible || est un objet sacre.
(3) II y a dans ce monde || si peu de voix et tant d'echos.
(4) Celui qui fait le bien en son temps || a travaille pour les
siMes.
(5) Pour vivre en paix avec les hommes, || il faut leur passer
bien des inegalites de caract^re. — ^Montesquieu.
(6) Pourvu qu'on sache la passion dominante de quelqu'un, ||
on est assure de lui plaire. — Pascal.
(7) Les plaisirs de la jeunesse reproduits par la memoire || sont
des ruines vues aux flambeaux. — Chateaubriand.
(8) Les gens d'esprit font beaucoup de fautes || parce qu'ils ne
croient jamais le monde aussi bete qu'il est.
(9) Ceux avec qui vous perdez votre temps, et qui vous le
derobent, || ne sont pas vos amis.
(10) Pour les ames de bonne volonte, || il n'est pas une minute
dans la vie qui n'ait son devoir.
(11) La mort n'est que le plus puissant acte de la vie, || car elle
enfante une vie sup^rieure. — Vergniaud.
204 MANUAL OF FRENCH PRONUNCIATION
(12) Plus on est honnete homme, || plus on a de la peine a
soupgonner les autres de ne I'etre pas. — Ciceron.
(13) Quand on jette les bonneurs k pleines mains, || beaucoup
d'indigents les ramassent et le merite se retire.
(14) Si vous voulez vous consoler, || pensez a tous les maux dont
vous etes exempts.
(15) La pauvrete coute plus cher que toutes les richesses, || car
on ne pent y arriver qu'cn donnant tout c.^ qu'on a. —
Vauvenargues.
(16) L'enfant devient pour ses parents, suivant I'education, ||
une recompense ou un chatiment.
(17) Parmi tant de gens a qui nous prodiguons le titre d'amis, ||
la plupart le sont juste assez pour nous dire bonjour.
(18) La simplicite de I'esprit et du coeur || est le meilleur moyen
de comprendre le vrai.
(19) L'homme qui combat pour la raison, pour la patrie, || ne
se tient pas si aisement pour vaincu. — ^Mirabeau.
(20) La masse de genie humain, par des alternatives de calme
et d'agitation, || marche toujours, quoique a pas lents, vers
une perfection plus grande. — ^Turgot.
II. The following sentence from Guy de Maupassant, in
Le Bonheur, is more complicated, but is an excellent specimen
of intonation :
" LTtalie, | ou chaque palais, | plein de chefs-d'oeuvre, |
est un chef-d'oeuvre | lui-meme, | ou le marbre, | le bois, | le
bronze, | le fer, | les metaux, | et les pierres | attestent | le genie
de l'homme, | ou les plus petits objets | anciens | qui trainent |
dans les vieilles maisons | revelent | ce divin souci | de la
grace, || est pour nous tous | la patrie sacree | que Ton aime, |
parce qu'elle nous montre | et nous prouve | I'effort, | la
grandeur, | la puissance, | et le triomphe | de I'intelligence
creatrice."
CHAPTER XXXVII
INTERROGATIONS
235. An interrogative sentence differs in intonation from
a declarative one in that it is uttered more or less with a
rising inflexion. As a rule, it commences with some word
or words essentially interrogative, such as pourquoi, comment,
est-ce, depuis quand, etc., or has the subject-pronoun placed
after the verb, as Vient-il ? Es-tu sur ? Frequently, how-
ever, such a sign is wanting, the speaker simply putting the
words into the declarative form and contenting himself with
the rise in the voice to give the interrogative meaning, as
Vous ne Vavez pas dit ? Vous venez ?
236. The main question is to determine what part of the
sentence should take the rising inflexion, as its place is not
always at the end of the sentence, but varies according to the
meaning. We have pointed out in the last chapter that
the usual declarative sentence has two parts, an upward one
implying that the statement is incomplete and that something
further is expected, and a downward one indicating com-
pleteness. The interrogative sentence really corresponds to
the first part only, i.e. it is an incomplete sentence. It has
a continuative effect, pointing to more coming, but does not
announce any more ; it indicates an expectation, but does
not satisfy it. The downward or concluding part of the
sentence, as it were, is wanting. Hence it is that an inter-
rogative sentence is usually uttered with a rising inflexion,
like the first part of an ordinary declarative statement ; and
the tonic word or syllable, i.e. the one which receives the
highest note of inflexion in the interrogative sentence, is the
same one that would receive it, if the sentence were put into
a declarative form, with both parts complete. To frame the
declarative form, one has only to supply an answer of some
205
2o6 MANUAL OF FRENCH PRONUNCIATION
kind to the interrogation, using identical or corresponding
words. Thus :
Interrogative ^
AUez-vous a I'etranger ? /
Est-il sans argent ? /
Quel chemin / a-t-il pris ?
Pourquoi / m'en voulez-vous ?
Pourquoi I'avez-vous puni ? /
Corresponding Declaration
Qui, c'est a I'etranger || que je
vais.
Non, il n'est pas sans argent, ||
mais sans amis.
II a pris le chemin || qui m^ne a
Dinard.
Voici pourquoi || je vous en veux.
Je I'ai puni || parce qu'il a
menti.
237. These explanations will enable the student to under-
stand the intonation of interrogative sentences. We now
give some practical rules, based on these explanations, but
subject to qualifications referred to afterwards :
(i) If there is a specially interrogative word in the
sentence, the rising inflexion is usually put on that
word, while the rest of the sentence, containing merely
complementary or subordinate ideas, is uttered on a lower
tone, either level or falling. Thus :
Comment / I'avez-vous trouve ? Pourquoi done / est-ce
qu'il a dit ga ? Jusqu'a quand / restera-t-il ici ? A quoi /
servent ces choses ? Sur quoi / donne votre fenetre ?
Qu'est / -ce que ce monde-1^ ?
(2) Similarly, when the interrogative clause is con-
tained in one part of the sentence only, the rising
inflexion comes at the end of that clause, as :
M'as-tu entendu / quand je t'ai appele } lEtiez-vous
fatigue? /dit le maitre. Que voulez-vous dire / avec ces
demi-mots ? Est-ce qu'il est arrive, / mon cher Henri ?
Est-ce clair maintenant, / monsieur } Et elle est sans fortune/,
vous dites ? Avez-vous oublie votre commission, / mon
ami ? As-tu dit cela, / malheureux ? On a frapp e : est-ce
toi, / Perrette ?
1 The sign (/), denoting a rising inflexion, is placed immediately after
the tonic word or syllable.
WORDS IN COMBINATION 207
(3) In all other cases, the rising inflexion usually
occurs at the end of the sentence, as :
Partirez-vous ? / Travaillerez-vous ? / Est-il dans la
chambre ? / Desirez-vous que je sorte d'ici ? / Vous ne le
saviez pas ? / Voulez-vous me rendre justice ? / Viendra-t-il
aujourd'hui ? / Tu veux me donner tant de mal ? /
In such cases there is sometimes a struggle between two
forces — ^the interrogation which requires the raising of the
note, and the end of the sentence which tends to lower it
as in ordinary declarative statements. Sometimes the latter
force succeeds, so that we may have, e.g., partirez-vous, with
vous somewhat lower than -rez, although not so low as it
would have been without the influence of the interrogation.
The question, as so put, is much less pressing than partirez-
vous ? I
238. The above three rules are subject to qualification.
Instead of the rising note being placed as mentioned, it may
be placed on some other word in the sentence, but in this
case the meaning of the interrogation becomes changed. The
tone being shifted, as it were, to this other word, the attention
is specially drawn to it. It becomes the main word in the
sentence, expressing the principal thing regarding which in-
formation is desired. It still corresponds to the word of
highest note in the declarative sentence, but the form of this
sentence becomes correspondingly altered. For example,
instead of the rising note being placed on the specifically
interrogative word (as in pourquoi / a-t-il dit ga?), it may be
placed at the end of the sentence (as pourquoi a-t-il dit ga? I),
but in the former case, where the tone is on pourquoi, it is
understood that some reason exists for the utterance, and
the questioner wishes to know what this reason was, whereas
in the latter case, where the tone is on ga, weight is rather
laid on what was uttered (why did he say that, and not some-
thing else ?). The former question would correspond to the
declarative statement voild pourquoi \\il a dit ga, while the
latter would suggest il a dit ga, \\ parce qu'il n'a pas pu dire
autre chose.
The meaning of an interrogation may thus be completely
changed by a change in the ordinary intonation. The
following are further examples of this :
2o8 MANUAL OF FRENCH PRONUNCIATION
Alors pourquoi te donner tant de peine ? / Pourquoi
n'avez-vous pas de temoins ? / De quoi s'occupent vos
voisins ? / Comment appelles-tu 9a ? / Est-ce que vous
voulez / le faire ? Que faites-vous / cet hiver ? Vous a-t-il
dit / qu'il etait mon pere ?
239. When in an interrogation there are two or more
parallel sentences, the last one takes the downward inflexion
( \ ), as :
Apprenez-vous la geographic, / ou I'histoire naturelle ? \
Est-ce pour 9a qu'il est parti, / ou pour une autre chose ? \
Apporte-t-on la lampe pour la mettre sous le boisseau, / ou
sous le lit ? / N'est-ce pas pour la mettre sur le chandelier ? \
Quel est, parmi les animaux, celui qui marche a quatre
pattes le matin, / a deux pattes a midi, / et a trois pattes le
soir ? \
The upward inflexion, however, is required on the last of the
parallel sentences, if it is merely in apposition to the pre-
ceding one — that is, if it has the same relative meaning, as,
cueille-t-on des raisins sur des epines, / ou des figues sur des
chardons ? / The reason for this is, that in the other cases,
the last clause represents the downward part of the sentence
when turned into a declarative form, whereas in this case the
downward part is wanting.
240. Although an interrogative sentence corresponds to
the first part of a declarative one, the two are not identical
in intonation. As a rule, an interrogative sentence com-
mences three or four notes higher, and in cases of excitement
or passion its highest note is sometimes as much as an octave
above the highest note of a declarative sentence. The rise
too is more sudden and rapid, much of it taking place on the
tonic syllable itself.
CHAPTER XXXVIII
EXCLAMATIONS
241. An exclamation or an exclamative sentence is an
abrupt, inverted, or elliptical utterance, arising from sudden
and intense emotion, as Maudit soil ce jour-ld ! Yilou que
vous etes ! Quelle idee I
We have seen that the characteristic of interrogations is
a rising inflexion on the tonic word, but the characteristic
of exclamations is the Emphatic Stress. The specially
significant word in the phrase or sentence, as in the examples
just given, receives a great access of strength. In particular,
the first syllable in it beginning with a consonant
is seized upon forcibly and uttered with unusual stress. The
consonant is more prolonged than under ordinary em-
phasis. The vowel does not remain on a uniform note,
but takes a slight rising inflexion (although not much, as
height does not necessarily play any part in exclamations) ;
and what is most important of aJl, the whole syllable,
including generally the consonant following, is pronounced
with great intensity, sometimes three or four times greater
than in ordinary emphasis. These features may vary more
or less according to the speaker and the circumstances, but
they constitute the essential characteristic of all exclamations.
Thus, in the sentence, vous etes un iripon, we have an
ordinary declarative sentence, uttered with the usual em-
phatic stress on fri-. But if we say, iripon que vous etes !
the sentence becomes an exclamation, and the syllable frip-
now takes all the features of duration, height, and intensity
just described. Further examples :
Si vous saviez ! Bete que tu es ! La canaille ! Miserable
que vous etes ! Bigre ! Gaffeuse ! Diantre ! Fichtre !
Gare ! Ciel ! Ferme !
242. We have pointed out (§ 217) that in ordinary emphasis,
0 209
210 MANUAL OF FRENCH PRONUNCIATION
a word commencing with a vowel may take the emphatic
stress on the first syllable, provided the word is isolated,
as it were. In exclamations, however, an initial vowel
never receives the full stress, the reason being that it
does not afford sufficient support for an intense effort of
voice. The main exclamative stress invariably falls on the
next syllable, the initial consonant of which serves to give
the support needed. This consonant is lengthened, and it
is the vowel following it which receives the increased height
and the full intensity. The contrast between the first syllable
and the second serves to enhance the effect. Examples :
Imposteur que tu es ! Intriguant qu'il est ! Imbeciles
que nous sommes ! Abruti ! Ivrogne I Assassin !
243. Where there are interjections {oh! ah 1 hah 1 etc.),
or specially exclamative words [que, comhien, comme, quel,
etc.), these usually receive the main effect of the voice, and
the rest of the sentence is pronounced as usual with pro-
gressive descent, as :
Oh I monsieur, que c'est joli ! Quels chameaux !
Comme je vous remercie !
The stress in such cases consists in a lengthening of all
the consonants in the word (or of its vowel, if there be no
consonant), and a considerable increase in height and in-
tensity. The first consonant or group of consonants of the
word following (unless this word be merely parenthetical) also
shares in the lengthening. Thus, in the examples just given,
the c' which follows que, the ch which follows quels, and the
/ which follows comme are lengthened. The reason for this
is that the word or expression following que, quel, etc., is
really as regards signification the most important in the
sentence ; and though robbed of its emphasis through the
shifting of the stress away from it, it naturally retains its
consonantal lengthening almost intact. Further examples :
Oh ! dit-il, que j'etais sot ! Ah ! vous vous etes trompe !
Que d(e) crimes ils ont commis ! Que d(e) fois je suis venu
ici ! Que d(e) services il m'a rendus ! Que c(e) souvenir
vous est agreable ! Quelle perfidie ! Quels fripons !
Quelle mauvaise affaire ! Quel beau jardin ! Combien
Ne^ etait br^ve |
WORDS IN COMBINATION 211
Frequently, however, the stress, instead of falling thus on
the interjections and specially exclamative words, falls on
the really important or emphatic word. This is particu-
larly the case if the speaker wishes to draw special attention
to this word, or set it in relief. Instead, therefore, of saying,
quel emhetement ! with the exclamative stress on quel, we may
say, quel emhetement, with the full effort of voice on the
syllable -bete-. Other examples :
Quel outrage, cette affaire ! 0 ciel ! Quelle impatience !
Quelle effronterie ! Oh ! le miserable ! Oh ! la canaille !
244. In those cases where the exclamative word is reduced
to its mere consonantal element through the elision of the
e mute, this element still takes the usual lengthening and
intensity, but instead of the vowel following the elision being
stressed, the full effort of the voice falls on the first syllable
having a rhythmic accent. The following are examples of
this kind of exclamation, which is very common :
0 ciel ! dit men oncle^ qu'il est absurde de parler des
grosses dents !
Oh ! mademoiselle, qu'il est difficile de faire ce que vous
dites !
Imbecile ! qu'il est dangereux d'agir ainsi !
Qu'il etait nigaud de croire toutes ces balivernes !
Qu'on est a plaindre quand on est pauvre !
There are many other varieties of exclamations or exclama-
tive sentences besides those to which reference has been made
in this chapter, but sufficient has been said to enable the
student to understand how to pronounce them.
CHAPTER XXXIX
EXPRESSION
245. Clearness of articulation, correct accent, right em-
phasis, proper grouping and intonation, suitable inflexion,
and all the other qualities already mentioned are necessary
for good French. But assuming that all these have been
acquired, something more is still needed, viz. Expression or
Sound-shading. As a rule, even fluent conversationalists or
public speakers, unless they have this soul-quality, will fail
to influence or please. The most finished speaker is the one
who has so mastered every shade of expression that he can
adapt his tone to the special subject. After all, words and
sentences are mere abstract and neutral symbols of human
thought, and it is the particular shading given to them which
reveals the soul of the speaker and gives life and interest
to what he says. Appropriate * shading ' may completely
change the signification of a sentence : it may impart to such
phrases as ga va hien, vous Vavez dit, je devrais, etc., an ironical
or doubtful meaning, the opposite of the literal one. It
is not a case merely of individual words, but also of clauses,
sentences, and even paragraphs. This being so, a good
speaker, by enlisting the power of expression, materially
assists his hearers to understand his subject, while a speaker
who neglects this makes intelligent hearing laborious and
unpleasant. Two violin players may perform the same piece,
both displaying equal command over the technical difficulties ;
yet the playing of the one may hold a listener spell-bound,
whilst the other performance scarcely serves to awaken
interest. In the one case the violinist enters into the play-
ing, infuses his spirit into it, and gives it life, while the other
merely produces dead sound. So with speaking, and par-
ticularly with French, which can assume the most varied
shades, there must be a union between the speaker and the
matter of his speech. He must realize what he says, and
212
WORDS IN COMBINATION 213
must put soul and heart into it. Ideas of lightness and
dignity, storm and calm, love and hate, etc., must be ex-
pressed by distinct alterations of voice. A practised speaker
would not in the same tone or manner describe soldiers
charging and children playing, or a meeting with friends
and a struggle with a burglar.
246. This expression or sound-shading is very marked in
French on account of the emotional character of the people.
It is impossible, of course, to lay down definite rules for it —
every person cannot have the same power of expression :
it must come from inspiration and experience. It is suffi-
cient here to draw attention to the matter, and to mention
the following points :
(i) The High Tone (denoted in phonetic script by r before
the sentence) rises above the usual key. As already pointed
out, it occurs in interrogation and exclamation. It is also
used to express astonishment, excitement, eagerness, elevated
and joyous feelings, impetuous and imp^llsive passion, invective,
and anything which renders the speech intense. Thus :
' Es-tu la } ' Auguste I ' Te voila deja ! I Assez^ tu me fais
mal !
It is also proper for stirring description or animated talk.
(2) The Middle Tone is the tone of habitual utterance
or address unmarked by any of the above feelings, and is
used in ordinary conversation, declarative statements, narration,
descriptive remarks, moral reflection, or calm reasoning. Thus :
Tu es la. C'etait a Paris. La pensee fait la grandeur de
I'homme. II demeure a Rennes.
(3) The Low Tone (denoted by i_) falls below the usual
speaking key, and is employed in expressing feelings deeper
than ordinary, such ^sfear, incredulity, secrecy, grief, sorrow,
solemnity, gloom, melancholy, contempt, irony, and concentrated
passion. Thus :
l^ah ! |_Quel idiot ! j^oyez tranquille ! ( Je tremble que
votre supercherie ne soit decouverte. | Je suis triste
aujourd'hui. ~
Similarly, a humble or supplicating request is uttered on a
214 MANUAL OF FRENCH PRONUNCIATION
low tone, while a summary refusal, on the other hand, takes a
high tone, as :
|Voulez-vous bien me tirer d'embarras ? 'Certainement non !
247. The simple words oui and non, according as they are
uttered, may express the most varied ideas. For example,
in the case of oui, the ordinary falling inflexion {oui \ ) denotes
affirmation, pure and simple (" Yes, that is so "), while a
stronger falHng one (oui \) denotes irony, indifference, doubt.
The ordinary rising inflexion [oui / ?) expresses simple inter-
rogation, while a stronger rising one (oui I/}) corresponds to
great surprise or astonishment. Indeed, the shades imparted
by the voice are so expressive that oui may become equiva-
lent to a negative, and non to an affirmative.
CHAPTER XL
RHYTHM AND EURHYTHMY
248. The rhythm of a sentence depends, among other
things, upon the recurrence of stressed syllables at regular or
fairly regular intervals. In this sense it belongs peculiarly to
poetry, where it reaches its maximum, but it is found more
or less in well-constructed prose. A person of rhythmic ear,
though using only the plainest phraseology, may make it flow
with poetic smoothness.
In French these stressed syllables are the final ones of stress-
groups, i.e. those which have the Tonic Accent. If the words
in a sentence are selected and arranged in such a way that
these tonic stresses occur with a measured or timed movement,
the sentence is rhythmical in the sense just mentioned. Many
French speakers do this instinctively to some extent, not
only in elevated stye and public speaking, but even in
ordinary conversation, and their sentences thus become more
effective and more pleasant to the ear. Though the stress-
groups may vary considerably in the number of their syllables,
a speaker can equalize them to a large extent by dwelling
longer on the shorter groups — ^in other words, by lengthening
out their syllables. But as this draws attention to them
somewhat by putting them in relief, it is necessary that the
meaning should admit of this being done.
249. The good rhythm of a sentence, however — what is
known as Eurhythmy — depends on more than mere regularity
of accent. In the fine arts, such as sculpture, architecture,
and painting, eurhythmy is the harmonious combination of
the various lines and parts, and the term when applied to
diction has a similar meaning. It has already been pointed
out in the chapter on Intonation that the ordinary French
sentence may be divided into two parts, the first having an
upward inflexion, and the second a downward one. Each
215
2i6 MANUAL OF FRENCH PRONUNCIATION
of these parts may consist of several stress-groups, tv/o
or more of which may be combined syntactically. It is
only when these groups and combinations of groups are
so arranged that those in the second part correspond
harmoniously to those in the first, that a sentence pos-
sesses eurhythmy. In other words, the whole movement of
the sentence from first to last must be strictly harmonious
in its various sections, modelled according to the ideas ex-
pressed, and producing the effect intended. As it is evident
that in uttering a sentence these groups and combinations
require to be correctly grasped by the speaker before
they can receive the proper intonation, or duration, the
necessity arises for dealing with the subject in a manual of
pronunciation.
250. To explain the matter more particularly, we cannot
do better than here give the famous introductory sentence
from Bossuet's oration on Henriette-Marie de France. This
sentence has frequently been quoted as a specimen of perfect
eurhythmy, and one feels on reading it that it possesses
wonderful equilibrium and harmony :
" Celui qui r^gne | dans les cieux, | et de qui reinvent | tous
les empires, | a qui seul | apoartient | la gloire, | la majeste, |
et rindependance, || est aussi le seul | qui se glorifie | de faire
la loi I aux rois, | et de leur donner, | quand il lui plait, | de
grandes | et de terribles | lemons." ^
It will be noticed that the sentence, which divides itself
after Vindependance, has perfect equilibrium, there being
exactly the same number of stress-groups, viz. nine, in each
of the two parts. But what is really of consequence, the
sentence is perfectly harmonious throughout, for the com-
binations of groups which are specially joined by syntax
correspond in the two parts. Thus, the first part contains
group-combinations of (2-f 2)+5, and the second 4+5. Even
the important words in the one part correspond by position
to similar or opposite words in the other. Thus, the word
wis occurs in the second part in the place where empires
1 In familiar speech, de grandes et terribles legons would form only one
stress-group, but Bossuet's elevated style, like that of verse, requires
an accent on grandes and on terribles. (See § 203.)
WORDS IN COMBINATION 217
appears in the first, and the word legons in the place where
Vindependance is found.
We here give other examples of the same type of eurhythmy,
in which both equilibrium and harmony are found, with the
various combinations of groups in figures after each example :
L'homme ideal^, | le poete divin, | le grand artiste || defie
seul I le temps [ et les revolutions. | — Renan. (3 || 3.)
La justice | et la verite | sont deux pointes | si subtiles, ||
que nos instruments | sont trop mousses | pour y toucher |
exactement. | — Pascal. (4 II 4-)
Mais si les lois | de I'etat | s'opposent | a son salut |
eternel, || Dieu | ebranlera | tout I'etat | pour I'affranchir | de
ces lois. I — BossuET. (2, 3 || 3, 2.)
En arrivant | a un certain carrefour | ou nous nous
separions | pour prendre | des directions | differentes, || nous
fumes frappes | a la fois | de I'attitude | contemplative | de
Jean-Frangois | les Bas-Bleus. | — Nodier. {2, 4 || 4, 2.)
Quand I'obstacle | etait surmonte^ | et que I'attelage j
reprenait | sa marche | egale | et solennelle, || le laboureur J
jetait un regard j de contentement j paternel | sur son enfant, |
qui se retournait | pour lui sourire. | — George Sand. {2, 5 ||
5; 2.)
251. Perfect equilibrium, however, .is not necessary for
good rhythm. Not many sentences, indeed, except in elevated
style, are of this kind, and a too frequent recurrence of them
would prove wearisome. The vast majority have an unequal
number of stress-groups in each part. Nevertheless, they
may be * well-rhythmed,' provided the harmony of the two
parts is maintained. By this is not meant that the two
parts should be similar in grammatical form, as in the
sentence. Pour convaincre, il suffit de parley a V esprit ; pour
persuader, il faut alter jusqu'au cceur. This would be what
is known as a ' balanced' sentence, but might not necessarily
be proper rhythm. For eurhythmy, as already indicated, the
combinations of stress-groups in the one part must
correspond to combinations in the other, i.e. these com-
binations must contain the same number of groups or
multiples of them. The four sentences which immediately
follow the one already quoted from Bossuet's exordium will
show the matter more clearly than several pages of theory :
2i8 MANUAL OF FRENCH PRONUNCIATION
Soit qu'il eUve | les trones, | soit qu'il les abalsse, | soit
qu'il communique | sa puissance | aux princes, | soit qu'il la
retire | a lui-meme, | et ne leur laisse | que leur propre faib-
lesse, II il leur apprend | leurs devoirs j d'une mani^re |
souveraine | et digne de lui. | (3, 3, 2, 2 || 2, 3.)
Car, en leur donnant | sa puissance, || il leur commande |
d'en user j comme il fait | lui-meme j pour le bien | du monde. |
(2 II 2, 2, 2.) ^ ^
Et il leur fait voir, | en la retirant, || que toute leur majeste j
est empruntee, j et que, pour etre assis | sur le trone, | ils
n'en sont pas moins j sous sa main | et sous son autorite j
supreme. | (2 || 2, 2, 2, 2.)
C'est ainsi | qu'il instruit | les princes, || non seulement |
par des discours | et par des paroles, j mais encore | par des
effets I et par des exemples. | (3 || 3, 3.)
In the first sentence, the upward part, which has ten stress-
groups, commences with two combinations of 3 and ends with
two of 2, while the downward part, which has exactly half
the number of groups, commences with one combination
of 2 and ends with one of 3, without having any other com-
bination to cause disharmony. In the second and third
sentences, the upward parts contain each a combination of 2,
while the downward parts have the same combinations in
a series, without there being anywhere a combination of 3
to spoil the harmony. In the fourth sentence, the upward
part, containing a combination of 3, is followed by a down-
ward part containing two of the same, but nowhere is there
a combination of 2 to disturb the rhythm.
We give below some further examples, constructed on
different types, but each presenting a specimen of excellent
rhythm. Some of these examples no doubt might be divided
up differently, as no two Frenchmen might utter a sentence
exactly the same way. The tonic stress might be suppressed
or added at some points, and even the separation line between
the two parts might be otherwise placed ; but it will be
found that the divisions here given are in consonance with
the ideas conveyed, and that, even if these divisions were
altered, the harmony as a rule would not be affected :
Si nous n'avions pas | de defauts, || nous ne prendrions
pas I tant de plaisir | a les remarquer | chez les autres. |
— La Rochefoucauld. (2 || 2, 2.)
WORDS IN COMBINATION 219
Paraitre | un moment, | jeter un eclat | doux et pro fond, |
mourir | trds jeune, || voil^ la vie | d'un Dieu. | — Renan.
(2, 2, 2 II 2.)
Son nom j qu'il a toujours \k la bouche, j ses myst^res |
qu'il traite | si divinement, || rendront j sa simplicite | toute-
puissante. | — Bossuet. (3, 3 || 3.)
Cette superbe puissance, | ennemie j de la raison, | qui se
plait I a la controler j et a la dominer, | pour montrer | com-
bien elle peut | en toutes choses, || a etabli | dans I'homme |
une seconde nature. | — Pascal. (3, 3, 3 || 3.)
Un enfant | de six | a sept ans, | beau j comme un ange, |
et les epaules | couvertes, j sur la blouse, | d'une peau
d'agneau, || marchait | dans le sillon j parall^le | a la charrue, j
et piquait | les flancs | des boeufs | avec une gaule j longue j
et leg^re, | armee | d'un aiguillon. | — George Sand. (3, 2, 4 ||
4, 3, 3, 2.)
En somme, j malgre les exactions | des gouverneurs | et
les violences | inseparables | d'un gouvernement | absolu, || le
monde, | sous bien | des rapports, | n'avait pas encore ete |
aussi heureux. j — ^Renan. (i, 2, 4 || i, 2, 2.)
252. Whenever the harmony referred to is lacking, there
is no eurhythmy in the proper sense of the term. In such
cases the rhythm is said to be discordant. It may not
necessarily, however, on this account be bad rhythm. It is
a departure from the normal, a contrast to what should be
expected, but for that very reason the speaker may produce
an effect with it which otherwise would be unfelt ; for it is
by departing from the usual rule and presenting some un-
expected combination of groups that he can generally arrest
attention and produce effect. Pascal's sentence, Le silence \
eternel \ de ces espaces \ infinis || m'effraie, | (2, 2 || i), shows a
complete break of harmony, but he thereby puts the last
group into special prominence and makes it speak more
forcibly. Without a frequent break of rhythmical harmony,
important ideas would not be set in relief, and sentences
would tend to become weak and monotonous. The rhythm,
though broken, is not faulty so long as it accords with the
ideas meant to be conveyed. We give two or three examples
of discordant rhythm, intended to produce effect :
Seul I il est assis | a la droite j de Dieu le P^re i| pour
I'etemite. | — Renan. (4 || i.)
220 MANUAL OF FRENCH PRONUNCIATION
Que si vous me demandez | comment tant de factions |
opposees, | et tant de sectes | incompatibles^ | qui se devaient
apparemment detruire | les unes les autres, | ont pu si
opiniatrement | conspirer ensemble | contre le trone royal^ ||
vous I'allez apprendre. | — Bossuet. (5, 2, 3 || i.)
LDrsqu'en voyageant j.dans la presqu'ile | armoricaine, | on
depasse | la region, | plus rapprochee | du continent, | ou se
prolonge | la physionomie gaie, \ mais commune, | de la
Normandie | et du Maine, | et qu'on entre | dans la veritable |
Bretagne, | dans celle qui merite | ce nom | par la langue |
et la race, || le plus brusque changement | se fait sentir |
tout a coup. I (3, 4, 5, 3, 4 II 3.)
253. As the object of discordant rhythm is to produce
effect, the speaker or writer should guard against the habit of
drifting into it in almost every sentence. The indiscriminate or
excessive use of it not only defeats its object, but keeps the
mind of the hearer or reader always tense, expectant, and
on the alert, and thereby produces mental fatigue. Some
writers seem to be so constituted that they can only think
by contrast or disproportion, with the result that their pages,
though beautiful otherwise, are filled with discordant sentences
of which the reader soon tires. To avoid such results, a
frequent return should be made to harmonious rhythm,
which is pleasant to the ear and affords rest to the mind.
Great pains should be taken to secure the art of good
rhythm, as the power possessed by it to make speaking
pleasant and effective is very great. Perhaps nowhere over
the range of elocutionary effects can the taste and skill of
the speaker be more fully displayed than here. No one, of
course, can speak by rule, but one can so master an art as to
exercise it unconscious of its rules.
254. We referred at the commencement to regularity of
stress. In French versification this regularity reaches its
maximum. The lines, according to traditional rule, usually
consist of an equal number of written syllables, but the
important point is that they have a fixed number of stress-
groups, equal or almost equal in length. It is not so much
the number of syllables as this regular distribution of accents
that is the fundamental principle in French poetry, as in
all poetry ; and to pronounce the lines correctly, one should
WORDS IN COMBINATION 221
be able to beat time regularly so that each beat falls on the
accented syllable in a group. The following from Corneille's
Horace will show this :
Que le courroux | du ciel, | allum4 | par mes voeux, |
Fasse pleuvoir | sur elle | un deluge | de feux ! |
Puisse-je | de mes yeux | y voir tomber | ce foudre, |
Voir ses maisons | en cendre, | et tes lauriers | en poudre, |
Voir le dernier | Remain | a son dernier | soupir, |
Moi seule | en etre cause, | et mourir | de plaisir. | ^
Poetry can only sound rhythmical, as it is meant to do,
if this regularity of stress is observed. Each stress-group,
of course, may not have the same number of syllables, any
more than a bar of music has the same number of notes,
but each group should have equal time, as each bar of music
has.
^ The emphatic stress has been omitted, for the sake of clearness.
INDEX OF WORD-ENDINGS
Note, — The figures refer to the sections, n = footnote.
-a, 38 (I)
-a, 38 (2)
-ab, 38 (I)
-abe, 38 (I)
-able, 38 (i), 41 (10),
186
-abre, 41 (10), 186
-ac, 38 (i)
-ace, 38 (i)
-ache, 38 (i)
-dche, 41 (i)
-achme, 38 (i)
-acle, 38 (i), 186
-acre, 38 (i), 186
-act, 38 (i), 151 (4)
-acte, 38 (i)
-ad, -ade, 38 (i)
-adre, 41 (10), 186
-aen, 91
-af, -afe, 38 (i)
-afle, 38 (i), 186
-afre, 41 (10), 186
-ag, -age, 38 (i)
-agne, 38 (i)
-ague, 38 (I)
-ai, 21 (3), 25 (4)
-aible, 25 (4), 186
-aiche, 25 {4)
-aide, 25 (4)
-aie, 21 (3), 25 (6),
69
-aient, 25 (6), 69
-aigle, 25 (4), 186
-aigne, 25 (4)
-aigre, 25 (4), 104 (4) a,
186
-ail, 38 (I), 41 (4), 78,
112
-aile, 25 (4)
-aille, 41 (4), 76, 112
-^im 97
-aime, 104 (4) 2 c, 25
(4)
-ain, 86 (4) n., 97
-aine, 25 (4), 24 (3),
104 (4) 2 c.
-air, 25 (4)
-aise, 25 (4)
-aison, 25 (4), 31 (i)
-aisse, 25 (4), 24 (3),
104 (4) 2 c
-aive, 25 (4)
-al, 38 (i)
-albe, 38 (i)
-aid, -aide, 38 (i)
-ale, 38 (I)
-ale, 41 (i)
-algue, 38 (i)
-alme, 38 (i)
-alpe, 38 (i)
-alque, 38 (i)
-alte, 38 (I)
-alve, 38 (i)
-am, 38 (i), 86 (4), 91
-ambe, 91
-amble, 91, 186
-ambre, 91, 186
-ame, 38 (i)
-ame, 41 (i)
-imes, 38 (2), 104 (4)
2 a
-amp, -ampe, 91
-ample, -ampre, 91,
186
-an, 38 (I), 86 (4), 91
-ance, -anche, 91
-ancre, 91, 186
-ande, 91
-andre, 91, 186
-ane, 38 (i)
-ang, 91
-ange, 91
-angle, 91, 186
-anque, 91
-ans, 129 (3) n.
-anse, 91
-ant, -ante, 91
-autre, 91, 186
-aon, 91
-ap, -appe, 38 (i)
-apre, 41 (i), 186
-apte, 38 (i)
-aque, 38 (i)
-ar, 41 (II), 38 (i)
-arbre, 38 (i), 186
-arc, -arche, 38 (i)
-ard, -arde, 38 (i)
-are, 41 (11), 38 (i)
-arge, 38 (i)
-argue, 38 (i)
-arle, 38 (i)
-arme, 38 (i)
-ame, 38 (i)
-arpe, 38 (i)
-arque, 38 (i)
-arre, 38 (i), 41 (11)
-arse, 38 (i)
-artre, 38 (i), 186
-arve, 38 (i)
-as, 38 (i), 41 (2), 129
(2)
-ase, 41 (8)
-asme, 38 (i), 224 (2)
-asque, 38 (i)
-asse, 38 (i), 41 (3)
-assion, 41 (9)
-aste, 38 (i)
-astre, 38 (i), 186
-at, 38 (I), 151 (2) (3)
-dt, 38 (2)
-ate, 38 (I)
-ates, 38 (2), 104 (4)
2 a
223
224 MANUAL OF FRENCH PRONUNCIATION
-ation, 41 (9)
-ec, 25 (2)
-atre, 38 (i), 186
-ece, 25 (i), 104 (4) 2 c
-atre, 41 (i), 186
-6che, 25 (i)
-au, 47 (7)
-ect, 25 (2), 151 (4)
-aube, 47 (7)
-ecte, 25 (2)
-auche, 47 (7)
-ectre, 25 (2), 186
-aude, 47 (7)
-ed, 25 (2), 154
-auffe, 47 (7)
-ede, 25 (I)
-auge, 47 (7)
-Mre, 25 (I), 104 (4) a,
-aule, 47 (7)
186
-auld, 115 (3)
-ee, 21 (i), 69, 104
-ault, 115 (3)
(2)
-aume, -aune, 47 (7)
-een, 99
-aupe, 47 (7)
-ef, 25 (2)
-auque, 47 (7)
-efe, 25 (i)
-aur, aure, 50 (3)
-efle, 25 (I), 186
-ause, -ausse, 47 (7)
-ege, 25 (i)
-aut, aute, 47 (7)
-^gle, 25 (I), 186
-auve, 47 (7)
-egme, 25 {2)
-aux, 47 (7), 159
-egne, 25 (i)
-ave, 38 (I)
-egre, 25 (i), 186
-avre, 41 (10), 186
-egue, 25 (i)
-ax, 38 (I), 159
-eie, 25 (6)
-axe, 38 (I)
-eige, 25 (2)
-ay, 25 (5)
-eil, 25 (2), 78, 112
-aye, 25 (5), 41 (7) n.,
-eille, 25 (2), 76, 112
69
-eim, 97
-az, 41 (8). 133
-ein, 86 (4) n., 97
-aze, 41 (8)
-eindre, 97, 186
-eine, 25 (2), 24 (3),
-b, 147
104 (4) 2 c
-ble, -bre, 186
-eize, 25 (2)
-el, 25 (2)
-ceps, 129 (2)
-ele, 25 (I), :i04 (4)
-ch, 136 (2), 156
-cher, 123
2 c
-elions, -eliez, 189 (i)
-ck, 156
-elle, 25 (2)
-cle, -ere, 186
-em, 25 (2), 86 (4), 91
-ct, 151 (4)
-cueil, 67 n.
-enable, embre, 91, 186
-erne, 25 (i), 24 (3),
104 (4) 2 c
-d, 154
-dre, 186
-emment, 38 (3)
-empe, 91
-en, 86 (4), 91, 99
-ence, 91
-e, 21 (I)
-ene, 25 (i), 24 (3), 104
-eau, -eaux, 47 (7)
(4)2C
-eb, 25 (2)
-enfle, 91, 186
-ebe, 25 (I)
-ens, 99, 129 (3) n.
-hh\e, 25 (i), 186
-ent, 91, no
-ebre, 25 (i), 104 (4) a,
-ente, 91
186
-entre, 91, 186
-epe, 25 (i)
-epre, 25 (i), 186
-epte, 25 (2)
-eque, 25 (i)
-er, 25 (2), 123, 209
(4) w.
-erai, 69
-erais, 69
-erbe, 25 (2)
-erce, -erche, 25 (2)
-ercle, 25 {2), 186
-erde, 25 {2)
-erdre, 25 (2), 186
-ere, 25 (i)
-erement, 25 (i), 104
(4) 3 n.
-erent, 25 (i)
-erf, 25 (2)
-erg, -erge, 25 (2)
-ergue, 25 (2)
-erie, 69
-erions, -eriez, 189 (i)
-CFme, -erne, 25 (2)
-erte, 25 (2)
-erve, 25 (2)
-erre, 25 (2)
-erse, 25 (2)
-es, 25 (I), 24 (i), 104
(4) 2 b, 129 (2)
-ese, 25 (I)
-esme, 25 (2), 224 (2)
-esque, 25 (2)
-esse, 25 (2), 104 (4)
2 c
-est, -este, 25 (2)
-estre, 25 (2), 186
-et, 25 (2), 151 (2) (3)
-^te, 25 (I), 104 (4)
2 c
-ete, 25 (i), 104 (4)
2 a
-eter, 188
-etre, 25 (i), 104 (4)
2 a, 186
-ette, 25 (2)
-eu, 64 (i)
-euble, 65, 186
-eue, 64 (i)
-euf, 65
-eugle, 65
-euil, 66, 78, 112
-euille, 66, 76, 112
INDEX OF WORD-ENDINGS
225
-eul, -eule, 65
-eumatique, 64 (3) n.
-€une, 65
-eur, 66
-cure, 62, 66
-eurer, -eurrcr, 104 {4)
-eus, 64 (2)
-euse, 64 (2)
-eut, 64 (i)
-eute, 64 (2)
-cutique, 64 (3) n.
-eutre, 64 (2), 186
-cuve, 66
-euvre, 65, 186
-eux, 64 i)
-6ve, 25 (i)
-ex, 25 (2), 159
-exe, 25 {2)
-exte, 25 {2)
-ey, 25 (5)
-eye, 25 (5)
-ez, 25 (2), 133
-eze, 25 (i)
-f, 125
-fle, -fre, 186
-g, 160, 163
-ger, 123
-gle, -gre, 186
-gue, 62, 69
-gueil, 67 n.
-i, 15
-ibe, 16
-ible, 16, 186
-ibre, 16, 186
-ic, -ice, -iche, 16
-icle, 16, 186
-ict, 16, 151 (4)
-icte, 16
-ide, 16
-idre, 16, 186
-ie, 16, 69, 104 (2)
-ien, 99
-lent, 69
-ier, 123 .
-if, -ife, 16
-ilie, -if re, i6, 186
-ige, 16
-igne, 16
-igre, 16, 186
-igue, 16
-il, 16, 115 (I)
-ile, 16
-ille, 16, 76, 112
-iltre, 16, 186
-im, 16, 86 (4), 97
-imbre, 97, 186
-ime, 16
-imes, 16, 104 (4) 2 a
-impe, 97
-in, 86 (4), 97, ^01
-ince, 97
-incre, 97, 186
-inct, 97, 151 (4)
-inde, 97
-indre, 97, 186
-ine, 16
-inge, 97
-ingle, -ingre, 97
-ingue, 97
-inmes, 97
-inrent, 97
-ins, 129 (3) n.
-insse, 97
-inte, 97
-intra, 97, 186
-inx, 97, 159
-ipe, 16
-iple, 1 6, 186
-ipse, 16
-ique, 16
-ir, -ire, 16
-is, 16, 104 (4) 2 b, 130
(2)
-ise, 16
-iser, 104 (4) 3 n.
-is me, 16, 224 (2)
-isque, 16
-isse, 16
-ist, -iste, 16
-istre, 16, 186
-it, 16, 151 (2)
-ite, 16
-ites, 16, 104 (4) 2 a
-itre, 16, 186
-ive, 16
-ivre, 16, 186
-ix, 16, 159
-ixe, 1 6
-ixte, 16
-k, 156
■1, 115
-las, 41 (2) n.
-lier, 189 (2)
-It, 151 (3)
-m, 140
-man, 86 (4) n.
-n, 140
-nas, 41 (2) n.
-nier, 189 (2)
-o, 47 (I)
-ob, -obe, 50 (i)
-oble, obre, 50 (i),
186
-oc, -oche, 50 (i)
-ocle, -ocre, 50 (i),
186
-ode, 50 (I)
-oe, 21
-oele, 41 (12)
-oeur, 66
-oeuvre, 66, 186
-ofe, 50 (I)
-ofie, -of re, 50 (i), 186
-og, oge, 50 (i)
-ogne, 50 (i)
-ogre, 50 (I), 186
-ogue, 50 (I)
-oi, 38 (4) ^
-oient, 38 (6), 41 (7). 69
-oiffe, 38 (4)
-oigne, 143
-oil, 112 «.
-oile, 38 (4)
-oin, 72 (4), 97
-oindre, 97, 186
-oine, 38 (4)
-ointe, 97
-oir, -oire, 38 (4)
-ois, 38 (4)
-oise, 41 (8) n.
-oit, -oite, 38 (4)
-oitre, 38 (4), 186
-oive, 38 (4)
-ol, -ole, 50 (i)
-ole, 47 (2)
226 MANUAL OF FRENCH PRONUNCIATION
-olte, 50 (i)
-olve, 50 (i)
-om, 50 (I), 86 (4),
95
-ombe, 93
-omble, -ombre, 93,
186
-ome, 50 (i), 47 (5)
-ome, 47 (2)
-omme, 47 (5)
-ompe, 93
-ompre, 93, 186
-on, 86 (4), 95, 50
(i)
-one, 93, 156
-onee, -onche, 93
-onde, 93
-ondre, 93, 186
-one, 50 (I), 47 (5)
-onfle, 93, 186
-onge, 93
-ongle, 93. 186
-onne, 47 (5)
-onque, 93
-ont, -onte, 93
-ontre, 93, i86
-onze, 93
-op, -ope, 50 (i)
-ople, -opre, 50 (i), 186
-oque, 50 (I)
-or, -ore, 50 (i)
-orbe, 50 (i)
-orche, 50 (i)
-orde, 50 (i)
-ordre, 50 (i), 186
-orge, 50 (i)
-orgne, 50 (i)
-orgue, 50 (I)
-orme, -ome, 50 (i)
-orque, 50 (i)
-orse, -orte, 50 (i)
-OS, 50 (I), 104 (4) 2 b,
129 (2)
-ose, 47 (4)
-oser, -osif, 47 (4)
-osion, osite, 47 (4)
-osition, 47 (4)
-osse, 50 (I), 47 (6)
-oste, 50 (i)
-ot, 47 (I)
-ote, 50 (i)
-ote, 47 (2)
-otion, 47 (3)1
-otre, 47 (2), 186
-ou, 56
-ouble, -oucle, 56, 186
-ouche, 56
-oude, 56
-oudre, 56, 186
-oue, 56
-ouffe, 56
-oufle, -oufre, 56, 186
-ouge, 56
-ougre, 56, 186
-ouil, 56, 78, 112
-ouille, 56, 76, 112
-ouin, 97
-ould, 56, 115 (3)
-oule, 56
-oult, 56, 115 (3)
-oupe, 56
-ouple, 56, 186
-our, 56
-ourbe, -ourde, 56
-ourse, -ourte, 56
-ous, 56, 104 (4) 2 b
-ouse, 56
-ouser, 104 (4) 3 n.
-ousse, 56
-ouste, -oute, 56
-outre, 56, 186
-ouve, 56
-ouvre, 56, 186
-oux, 56
-ouze, 56
-ove, 50 (i)
-ove, 47 (2)
-ovre, 50 (i), 186
-oxe, 50 (i)
-oy, 38 (5)
-oz, 56, 133
-P, 147
-pie, pre, 186
-pt, 151 (4)
-q, 156
-r, 123
-racle, 41 (10)
-ras, 41 (2) w.
-rer, 104 (4) 3 n.
-Tie, i«6
-roi, 41 (5)
-roie, 41 (7)
-roy, 41 (6)
-rs, 129 (3) n.
-s, 129, 209 (4) n.
-sme, 224 {2)
-St, 151 (3)
-t, 151
-th, 148
-tas, 41 (2) n.
-tie, 150 (5)
-tie, 150 (4)
-tier, -ti^re, 150 (4)
-tidme, 150 (4)
-tions, -tiez, 150 (5)
-tre, 186
-u, 61
-ube, 61
-uble, -ubre, 61, 186
-uc, 61
-uce, -uche, 61
-ucre, 61, 186
-ud, -ude, 61
-ue, 61, 69, 104 (2)
-ue, 62
-ueil, 66, 78, 67 n.
-ueille, 66, 76, 67 n.
-ufle, 61, 186
-uge, 61
-ugle, 61, 186
-ugne, 61
-ul, -ule, 61
-ulte, 61
-um, 50 (4), 86 (4), 95,
100
-umble, 100
-ume, 61
-un, 95, 100, loi
-une, 61
-unte, 100
-upe, 61
-uple, 61, 186
-uque, 61
-urent, 61
-urge, 61
-ume, 61
-urque, 61
INDEX OF WORD-ENDINGS
22^
-ur, -ure, 6i
-us, 6i, 104 (4)
129 (2)
-use, 61
-usque, 61
-usse, 61
-uste, 61
-ustre, 61, 186
2 b,
-ut, 61, 151 (3)
-ute, 61
-uve, 61
-ux, 61, 159
-uz, 61, 133
-vre, 186
-X, 130, 159
-yen, 99
-yer, 123
-ym, yn, 97
-ypte, 16
-yrte, 16
-yx, 16, 159
-z, 133
p ♦
INDEX OF
PRINCIPAL WORDS CITED
Note. — ^The numbers refer to the sections, m. = footnote.
Where a word is not given here, students are recommended to look
in the index of word-endings.
ab-, 224 (i)
abbaye, 21
abc6s, 221
abime, 16
abreuve, 68
abreuver, 63
Achille, 114
Adam, 86 (4) n.
adroite, 104 {3)
afflux, 159
Afghan, 223
aigu-, 161
aigue, 62
aiguille, 161
aiguiser, 161
Aix, 159
Aix-les -Bains, 130
Aix-la-Chapelle, 130
Alfred, 154
almanach, 136 (2)
Als-, 224 (4)
ambesas, 129 (i)
ambiguite, 161
amict, 151 (4)
Amiens, 129 (3) n.
anecdote, 221
Anne, 41 (12)
anspect, 151 (4)
antienne, 150 (6)
aout, 57
appendice, 99
aprds, 209 (i)
aquarelle, 157
aquarium, 157
aquatique, 157
arc-boutant, 156
arc-doubleau, 156
archi-, 136
as, I, 129 (i)
aspect, 151 (4), 209
(4) w.
assieds, 21
asthme, 152
Auch, 50 (3), 136
au-dessus, 223
augmenter, 50 (3)
aulnaie, 115 (2)
aulne, 115 (2)
aulx, 115 (2)
aumone, 47 (2), 50 (3)
aumonier, 50 (3)
automne, 86 (3)
autrui, 80 n.
Auxerre, 50 {3), 130
Auxois, 50 (3), 130
Auxonne, 50 (3), 130
avant-hier, 74
av^nement, 30 (2) n.
babil, 115 (i)
babord, 41 (i)
bacille, 114
Bagdad, 154
balandras, 129 (2)
bals-, 224 (4)
banc, 156
bapt^me, 146
baptismal, 146 n.
Bavarois, 38 (4)
bayadere, 25 (5) n.
Bayard, 25 {5) n.
bayer, 25 {5) n.
Bayeux, 25 (5) n.
Bayonne, 25 (5) n.
Beatrix, 130
beaucoup, 205
bec-d'dne, 156
Belsunce, 99, 224 (4)
benzine, 99
Berlioz, 133
beta, 38 (I)
beugle, beugler, 68
bienfaisance, 70 (4)
bienfaisant, 70 (4)
billevesee, 114
billion, 114
bis, 129 (i)
Biscaye, 25 (5) n.
blanc, 156
blason, 41 (8)
boa, 50 (2)
bobo, 51 (2)
boeuf, 68, 125
boeuf-gras, 125
boeufs, 68, 125
bois, 41 (5)
boite, 41 (5)
boueux, 72 n.
Bourg, 163 M.
bourg, 163
bourgmestre, 163 «
bras, 38 (i)
Brest, 151 (i)
broc, 156
bruit, 80 n.
brut, 151 (i)
Bruxelles, 130
bruyamment, 81
INDEX OF PRINCIPAL WORDS CITED
229
bruyant, 81
bruy^re, 81
buffleterie, 33
but, 151 (i)
Buxy, 130
C-, 156
Caen, 69, 91
Calvados, 50 (i)
calville, 114
caoutchouc, 156
capillaire, 114
Cavaignac, 143
Cayenne, 25 (5) n.
ce, 185
celui, 116
cens, 129 (i)
centiare, 150 (6)
cent, 170
cep, 147
ce que, 185
cerf dix-cors, 125
cerf- volant, 125
cerfs, 125
ch-, 136 (I) (3), 156
Champaigne, 143
chaos, 129 (2)
chauve-souris, 223
chef-d'oeuvre, 125
cheptel, 146
Cherbourg, 163
chevecier, 33
chocolat, 41 (12)
chouette, 38 (7)
Christ, 151 (2)
chut, 62, 151 (i)
Cid, 154
cigue, 62
cinq, 168
Cinq-Mars, 129 (i)
cipaye, 25 (5) n.
circonspect, 151 (4)
clame, 41 (12)
Claretie, 150 (6)
clavecin, 223
clef, 21, 125
clerc, 156
climat, 41 (12)
club, 147
CO-, 50 (2)
cobaye, 25 (5) n.
COCO, 51 (2)
codicille, 114
combinaison, 31 (i)
compte, 146
construire, 80 n.
corinthien, 150 (6)
corps, 146
Cortez, 133
cotignac, 156
couenne, 38 (7)
couctte, 38 (7)
couler, 56, 104 (4) c
crabe, 41 (12)
Craon, 91
Craonne, 91
cr6neler, 30 (2)
cresson, 70 (5)
eric, 156
croc, 156
croup, 147
crucifix, 159
Crusoe, 21
cul, 115 (2)
Cyrille, 114
daim, 86 (4) n.
dam, 86 (4) n.
damas, 129 (2)
Damiens, 129 (3) n.
damne, 41 (12), 86 (4)
Daphnis, 16
David, 154
d6jeuner, 64 (3), 123
de le, 181
d6mener, 30 (2)
de ne, 181
depuis, 209 (i)
des-, 31 (2), 70 (5), 128
d6s-, 132
Des-, 128
desu6tude, 131 w.
d6truire, 80 n.
deut^ronome, 64 (3) n.
deux, 170
diapason, 41 (8)
diner, 123
dis-, 224 (i)
distiller, 114
district, 151 (4)
dix, 130, 168
dix-huit, 170
dodo, 51 (2)
doigt, 162
doigtcr, 162
dom, 86 (4) n.
dompter, 146
done, 156
dosse, 47 (6)
dot, 151 (i)
droite, 104 (3)
duo, 80 n.
Duplcix, 159
eaux, I
dchec, 156
Echelon, 30 (2)
6chevel6, 188
6crouler, 56, 104 (4) c
eczdma, 223
eff-, 31 (3)
dlever, 30 (2)
em-, 86 (2)
6meraude, 30 (2)
empoigner, 143
en-, 86 (2)
encens, 129 (i) n.
encoignure, 143
ensevelir, 188
6phod, 154
6pizootie, 150 (6)
6quateur, 157
dquestre, 157
Equitation, 157
es-, 31 (2)
6s, 129 (i)
esclave, 41 (12)
escroc, 156
espace, 41 (12)
essaim, 86 (4) n.
est, 151 (i)
estomac, 156
et, 211 (6)
6taim, 86 (4) n.
6tioler, 150 (6)
6troite, 104 {3)
eu, 62
eu-, 64 (4)
Eugene, 64 (4)
Eug6nie, 64 (4)
eiimes, 62
eurent, 62
Europe, 64 (4)
eus, 62
cut, 62
eiites, 62
230 MANUAL OF FRENCH PRONUNCIATION
evenement, 30 (2) n,
ex-, 31 (3), 158
exact, 151 (4)
examen, 86 (4) n., 99
exempt, 146
exemption, 146 n.
fa, 38 (I)
faim, 86 (4) n.
fais, 21
faisable, 70 (4)
faisant, 70 (4)
faiseur, 70 (4)
faisons, 70 (4)
fait, 151 (i)
faon, 91
faonner, 91
farouche, 136
fat, 151 (I)
faubourg, 163
fayard, 25 (5) n.
femme, 38 (3)
Feroe, 21
Jeudiste, 64 (3) n.
fez, 133
fils, I, 115 (2), 129 (i)
fimes, 16
fites, 16
flamme, 41 (12)
flanc, 156
fieur-de-lis, 129 (i)
flux, 159
Foch, 136 {2)
foi, 41 (5)
fort, 211
fosse, 47 (6)
fossette, 51 (i) n.
fossile, 51 (i) n.
fossoyer, 51 (i) n.
fossoyeur, 51 (i) n.
fouet, 38 (7)
franc, 156
froide, 104 (3)
froisse, 104 (3)
fruit, 80 n.
gageure, 62, 138
gagne, 41 (12)
gai, 21
galimatias, 150 (6)
garantie, 150 (6)
gars, 41 (12), 123
gaz, 133
geai, 21
Genevois, 33
gens, 129 (i)
gentilhomme, 112
gibeci^re, 223
Gille, 114
gingas, 129 (2)
girouette, 38 (7)
gite, 16
gn-, 144
goeland, 38 (7)
goelette, 38 (7)
goemon, 38 (7)
gogo, 51 (2)
gong, 163
Goths, 151 (3) n.
gouter, 123
granit, 151 (3)
gratis, 16
gresil, 112, 115 (i)
gril, 115 (i)
grog, 163
grosse, 47 (6)
group, 147
gruyere, 81
gu-, 161
Guadeloupe, 161
guano, 161
Guise, 161
Guyane, 161
Guyon, 161
h-, 165
hanap, 147
handicap, 147
hareng, 163
haut-de-chausses, 223
helas, 129 (i)
hennir, 38 (2)
heros, 167
Herschel, 134
hiatus, 74
hier, 74
hier-, 74
Hongrois, 38 (4)
hosanna, 47 (4)
houp, 147
huit, 168
hyene, 74
Hypatie, 150 (6)
ichneumon, 64 (3) n.
11, 116
ill-, 114, 172 (4)
ils, 116
imm-, 172 (4)
immangeable, 86 (2)
immanquable, 86 (2)
impromptu, 146 n.
inn-, 172 (4)
instinct, 151 (4)
instruire, 80 n.
indemnity, 38 (2)
indomp table, 146 n.
instiller, 114
iris, 16
irr-, 172 {4)
isthme, 152
jaconas, 129 (2)
Jacques, 41 (12)
jadis, 129 (i)
jalap, 147
Jamaique, 16
Jean, 91
Jeanne, 41 (12)
je le, 180
je me, 180
je ne, 180
Jesus, 129 (3) n.
J6sus Christ, 151 (2)
jeter, 223
jeudi, 64 (3)
jeun {k), 100
jeune, 64 (3)
jonc, 156
jouions, jouiez, 72 «.
jouet, 38 (7)
joug, 163
julep, 147
jungle, 95
junte, 95
k-, 155
la, 38 (I)
lacs, I, 156
la-dessus, 223
lampas, 129 {2)
Laonnais, 91
laps, 129 (i)
larynx, 159
lasting, 163
Lausanne, 47 {7)
INDEX OF PRINCIPAL WORDS CITED
231
Law, 130
le, 189 (4)
Leclerc, 156
legs, 162
les-, 128
lichen, 136 (3)
Liebig, 163
Lille, 114
lilliputien, 114
linceul, 78
lingual, 161
linguiste, 161
lis, 129 (I)
lolo, 51 (2)
long, 205
Longwy, 162
loquace, 157
losange, 47 (4)
lumbago, 95
lut, luth, 151 (I)
Lycaon, 91
Mabille, 114
Machiavel, 136 (3)
mais, 209 (i)
maison, 31 (i)
mademoiselle, 154 n.
malfaisant, 70 (4)
malfaisance, 70 (4)
mangeure, 62, 138
raanne, 41 (12)
marc, 156
Marc, 156
marqueterie, 33
Mars, 129 (i)
mat, 151 (i)
mat, 151 (i)
matras, 129 (2)
Mauclerc, 156
mauvais, 50 (3)
mayonnaise, 25 (5) n.
maxillaire, 114
medecin, 221
menthol, 99
mes-, 31 (2), 128
mes-, 131
m6tis, 129 (i)
Metz, 129 (3) n.
meule, 64 (3)
Meung, 100
meunier, 64 (3)
Michel- Ange, 136 (3)
mil, 112, 115 (i)
mill-, 114
mille, 114
Millet, 114
miroir, 38 (4)
moelle, 38 (7)
moelleux, 38 (7)
moellon, 38 (7)
moeurs, 129 (i)
moignon, 143
mois, 41 (5)
momie, 47 (8)
monsieur, 64 (i), 70
(i), 123, 224 (3)
Mont-, 152
Montaigne, 143
mosaique, 47 (4)
mouette, 38 (7)
mousqueterie, 33
myosotis, 47 (4)
myrtille, 114
naif, 16
nenni, 38 (2)
nerf, 125
nerf-de-boeuf, 125
net, 151 (i)
neuf, 168, 205
Neufchateau, 125
Neufchatel, 125
neume, 64 (3)
Nimes, 16
Noel, 25 (i)
noix, 41 (5)
nom, 86 (4) n.
non, 247
nord-est, 211
nord-ouest, 211
noueux, 72 n.
Nyons, 129 (3) n.
oasis, 50 (2)
ob-, 224 (i)
obus, 47 (8)
odeur, 47 (8)
odieux, 47 (8)
oe-, 21
ceil, 112
oeuf, oeufs, 68, 125
oignon, 143
oisivete, 223
orchestra, 136 (3)
orchis, 136 (3)
oreiller, 123
ortie, 150 (6)
ortier, 150 (6)
OS, 129 (i)
osciller, 114
Osiris, 16
ouate, 211 (5)
ouest, 151 (i)
oui, 211 (5), 247
ours, 129 (i)
paon, 91
paonneau, 91
paonner, 91
Parfaict, 151 (4)
parfum, 86 (4) w.
parisis, 129 (i)
paroisse, 104 (3)
parler, 123
parqueterie, 33
partie, 150 (6)
pat, 151 (I)
pataques,24(i), 129(1)
Paul, 50 (3)
Paule, 50 (3)
pays, 21, 77
pentateuque, 64 (3),
99
pente-, 99
pentecote, 99
perdrix, 159
peu, 68
peut-dtre, 70 (2)
Pharaon, 91
philosophe, 47 (4)
pluie, 80 n.
plus, 129 (I)
poids, 41 (5)
poignant, 143
poignard, 143
poigne, poign6e, 143
pois, 41 (5)
Poitiers, 150 (4)
poix, 41 (5)
pore, 156
post-dater, 152
post-scriptum, 152
ponding, 163
poulailler, 123
pouls, 115 (2)
pre-, 131
232 MANUAL OF FRENCH PRONUNCIATION
preposition, 47 (4)
pres-, 31 (2)
presanctifier, 131 n.
pres6ance, 131 m,
prfeupposer, 131 «.
po61e, 41 (12)
prix, 159
prompt, 146
pros-, 47 (4)
pseudonyme, 64 (3)
punch, 136
pupille, 114
pusillanime, 114
Pythie, 150 (6)
qu-, 157
quad-, 157
quadrille, 157 w.
quai, 21
quaker, 157
quand, 209 (i)
quartz, 133, 157
quasimodo, 157
quatre, 209 (5)
que je, 181
que le, 181
quelque, 116
questure, 157
qui6tisme, 157
Quintilien, 157
quintuple, 157
quorum, 157
raccroc, 156
rail, 41 {4)
raisin, raison, 31 (i)
rang, 205, 209 (4) n.
re-, 184
re-, 131
redevenir, 188
reflux, 159
reine-claude, 164
rejeton, 223
rem-, 86 (2)
repartie, 150 (6)
reps, 129 (i)
requiem, 157
res-, 31 (2), 70 (5)
resection, 132 w,
r6s6quer, 132 n.
respect, 151 (4), 209
(4)«.
ressemeler, 188
Retz, 129 (3) n.
rez-de-chauss6e, 223
rhumatisme, 224 (2)
Richelieu, 189
rien, 189
rit, 151 (I)
rob, 147
rococo, 51 (2)
Rodez, 133
roide, 104 (3)
romps, rompt, 146
rotie, 150 (6)
rouennerie, 38 (7)
rouet, 38 (7)
rouler, 56, 104 (4) c
rumb, 147
ruolz, 133
Saens, 69
sais, 21
Saint-Brieuc, 156
Saint-Ouen, 99
Saint-Marc, 156
salep, 147
sang, 205, 209 (4) n.
sangsue, 162
Saone, 41 (2)
saoul, 56, 115 (2)
Sarmatie, 150 (6)
sassafras, 129 (2)
sauf, 50 (3)
sauvetage, 223
savetier, 221
SC-, 130
sceptique, 130
scille, 114
scintnier, 114
scottisch, 134
sculpter, 146
scythe, 16
se, 185
second, 164, 222
Seidlitz, 133
selon, 209 (i)
senefon, 33
sens, 129 (i)
sept, 146, 168
serfs, 125
signet, 162
Sillery, 114
Sinai, 16
Sind, 154
six, 130, 168
six-huit, 170 n.
sm-, 224 (2)
Soissons, 129 (3) n.
soit, 151 (i)
soixante, 130
solennel, 38 (3)
solennit6, 38 (3)
sortie, 150 (6)
sotie, 150 (6)
soucoupe, 70 (3)
souhaite, 38 {7)
soul, 56, 115 (2)
souper, 123
sphinx, 159
square, 157
Stael, 69
Strasbourg, 221
sub-, 224 (i)
subsister, 224 (i)
sud, 154
Suez, 133
sur, 121
sus, 129 (i)
susdit, 223
suspect, 151 (4)
sympathie, 150 (6)
tabac, 156
talmud, 154
te le, 181
temps, I, 146
tenir, 99
terroir, 38 (4)
t6tras, 129 (2)
teuton, 64 (3) n.
th-, 148
thym, 86 (4) n.
ti-, 150 (I)
tiers-6tat, 211
tiroir, 38 (4)
titiller, 114
toit, 41 (5)
tournevis, 129 (i)
tous, 56, 104 (4) c,
129 (i)
tranquille, 114
trans-, 131, 224 (i)
transept, 131 «.
trans vaser, 221
Transylvanie, 131 «.
INDEX OF PRINCIPAL WORDS CITED
233
trois, 170
tronc, 156
trop, 47 (i), 205
Troyes, 38 (5)
truie, truite, 80 n.
tu, 60 n.
tuyau, 81
tuyere, 81
ubiquit6, 157
uhlan, 211 (5)
un, 170 n., 211 (4)
us, 206
ut, 151 (i)
vaciller, 114
vainc, 156
vais, 21
vaudeville, 114
venir, 99
vergeure, 62, 138
verticille, 114 '
veule, 104 (4) c
vill-, 114
ville, 114
vingt, 162, 169
vis, 129 (i)
voix, 41 (5)
volontiers, 123, 150 (4)
W-, 73 (5), 126
wagon, 126
wallon, 72 (5)
wh-, 72 (5)
whig, 163
X-, 158 n.
Xer^s, 158
Ximen^, 158
y-, 74
yacht, 151 (3), 211 (5)
yankee, 211 (5)
yatagan, 211 (5)
yeuse, 74 ^
yod, 154
zend, 154
zig-zag, 163
zinc, 156, 65
zut, 151 (I)
INDEX OF SUBJECTS
Note. — The figures refer to the sections.
Phonetic symbols are enclosed in square brackets.
Prefixes will be found in List of Words.
«. = footnote.
a, close, 34- ; open, 39- ; close
and open, 34, 35, 42 ; un-
stressed, 37, 43 ; final,
38 (i) ; in derivatives, 43 (i) ;
before mute consonant,
38 (i) ; in closed termina-
tions, 38 (i) ; in non-final
syllables, 38 (i) ; before s,
41 {2) (3) ; before z-sound,
41 (8), 43 (3) ; before final
r, 41 (11) ; before medial
r, 43 (4) ; middle or inter-
mediate, 43 (4)
[a], how pronounced, 4, 8, 34, 35 ;
in ordinary spelling, 38 ;
duration, 37, 104 {4)
[a], how pronounced, 4, 8, 34,
35, 39 ; long and short, 40 ;
in ordinary spelling, 41 ; be-
coming [a], 43 (2) ; duration,
104 (3)
[a] how pronounced, 4, 91 ; in
ordinary spelling, 91 ; dis-
tinguished from 5, 94
^, 37, 38 (2)
^, 37, 38 (2), 40, 41
accent, tonic, 16 n., 193- ; le in,
104 (i) ; influence on length
of vowel, 104 (3) (4) ; in
isolated words, 194 ; in stress-
groups, 196 ; in elevated
speech, 203 ; preceded by
emphatic accent, 219
accent, emphatic, 214, 241
ae, 36, 96
ai, 21 (3), 24, 25 (4), 31 (I), 70 (4)
ai, 24, 25 (I)
aie, 24, 25 (6)
234
American English, 87 (3), 226
apris, in liaison, 209 n.
alphabet, phonetic, 3
aon, 91
aspect, in liaison, 209 (4) n.
aspirate, h, list of words, 167 ; no
elision before, 189 (3)
assimilation, vocalic, 29, 51 (2),
68 ; consonantal, 106, 221
au, 44 ; unstressed, 47 (7), before
r, 50 (3)
ay, 24, 25 (5)
aye, 24
b, final, 147
[b], how pronounced, 145 ; changed
to m, 88 (2) ; lengthening
consonant, 104 (4)
Bibliography, 240
bi-labial consonants, 107
breath, how to take, 234
breath-groups, 199
c, final, no, 156; for [s], 130;
for [k], 156
9, 130
caduc, e, 69
ch, sounded [J], 134 ; sounded [k],
136
circumflex, influence on length,
104
classical words, 86 (4), 104 (4) 2 (b),
129 (2)
clearness in vowel sounds, 13
compound words, 132
conditionals and futures, e mute
INDEX OF SUBJECTS
235
consonants, defined, 7, 105 ;
lengthening, 104 (4) ; voiced,
106 ; unvoiced, 106 ; division
of, 107 ; final, no, 170 n.,
204 ; mute following r in
liaison, 211 (i) ; phonetic,
3 ; nasal, 107 ; double, 171- ;
Law of Three, 176, excep-
tions to, 178
consonantal group, followed by
e mute, 186
continues, 179 (i)
cuir, 208 n.
d, how pronounced, 153 ; final,
154 ; in liaison, 205
[d], lengthening consonant, 104
(4) ; changed into n, 88 (2) ;
doubled, 172 (4)
dates, 168
dental consonants, 107
depuis, in liaison, 209 n.
derivatives, how pronounced, 27,
41 (3), 43 (I), 51 (i)
diphthongs, 12
disaccentuation, 202
division of syllables, 5 ; in verse,
75, 86 n.
double consonants, 171
downward part of sentences, 232
duration of sounds, 103-
e, close or open, 4, 8, 17 ; in
closed syllables, 25 (2) ;
before a double consonant,
25 (2) ; sounded [a], 37,
38 (3) ; unstressed, 27- ;
becoming mute, 33 ; middle,
28- ; in derivatives, 27
e mute, 17, 31 (2) n. ; becoming
W. 33 I becoming [oe], 70 ;
when not pronounced, 69 ;
in ordinary spelling, 69 ; in
initial syllables, 179 ; in fixed
groups, 180 ; before re-, 184 ;
after s-sound, 185; final,
after consonantal group, 186 ;
cases where never elided,
189 ; in verse, 191 ; inserted
wrongly, 192
[e], how pronounced, 4, 8, 18, 19 ;
duration, 104 (i) ; short
usually, 18, 20; tense, 19;
in ordinary spelling, 21 ; sub-
stituted for e mute, 70
[e], how pronounced, 4, 8, 22,
23 ; followed by r, 24 n. ;
duration, 24, 104 (4) ; in
ordinary spelling, 25 ; final,
26
[g], how pronounced, 4, 96 ; in
ordinary spelling, 97
[a], how pronounced, 4, 69 ;
duration, 104 (i) ; final,
after vowel, 104 (2) ; elision
of, 175
[0], how pronounced, 4, 63 ; long
and short, 64 ; in ordinary
spelling, 64 ; duration, 104 (3)
e, unstressed, followed by r, 30
(i) ; before a consonant
followed by silent e mute,
30 (2) ; when representing
h in stem, 30 (3) ; when initial,
30 (3)
^, 24, 25 (I)
e, 24, 25 (i)
e, 25 (i), 62 n.
eau, 47 (7)
ei, 24, 25 (I) ; eie, 25 (6)
elision, 175 ; in pronoun le, 70 n. ;
none before numerals, 170,
189 (3)
emotional words, emphasis in,
315
emphasis, [e] and [9] in, 104 (i)
emphatic accent, see accent
enumeration, emphasis in, 220
equilibrium in speech, 250-
-er, in liaison, 209 (4)
es, 31 (2)
et, no liaison with, 211 (6)
eu, close, 63-
eu, open, 65-
Eurhythmy, 248-
exclamations, 241-
explosives, 107, 145
expressions, 245-
ey, 24, 25 (5)
f, final, no, 125 ; in liaison, 205
[f], how pronounced, 124
final vowels, 104 (2) ; followed by
e mute, 104 (2)
236 MANUAL OF FRENCH PRONUNCIATION
final consonants, no, 170 n., 204
fixed groups, 180, 185
fixed words, 188
foreign words, 86 (4), 99 (3), 147,
151 (3), I54> 156, 157, 160,
161, 163
fort, in liaison, 211 (i)
fricatives, 107, 124-
fundamental vowel-sounds, 4
futures and conditionals, e mute in,
188, 189
g, medial, 162, 164 ; final, 163,
164 ; in liaison, 205
[g], 3; how pronounced, 160;
lengthening consonant, 104
(4)
ge, before a, o, u, 138
glottal consonant (h), 107, 165
gn [ji], 141
gn [gn], 145
gradation, emphasis in, 220
grasseyer, 119
Greek and Latin words, 86 (3),
99 (3), 104 (4) 2 (b), 129 (2),
167
Greek w, 50 (i) n.
group, final consonantal, followed
by e mute, 186
group, stress, 16, 104 (4) 3 ; in
elevated speech, 203 ; liaison
in, 207
group, breath, 199
gu-, 161
h, 165- ; aspirate, list of words,
167 ; no elision before, 189 (3) ;
no liaison before, 211 (3)
half-long sounds, 40 n. ; 104 (3)
hard consonants, 106
harmony in speech, 250-
high tone, 246 (i)
homonyms, i, 41 (5), 92, 98
humming, 88 (i)
i, at end of small unstressed words,
79
[i] how pronounced, 4, 8, 15 ;
duration, 16, 104 (4)
[i] (open), 15
[J], 3, 7, 73 ; initial, 74 ; medial,
75 ; final, 78 ; sometimes
inserted, 77 ; lengthening
consonant, 104 (4)
-ign-, 143
-il, how pronounced, 78, 112,
114 (I)
-ill, how pronounced, 76, 112, 113,
114
indeterminate vowel sound, 4, 69-
inflection, 237-
initial syllable, e mute in, 179
intermediate e, 28
intermediate o, 51 (3)
intermediate a, 43 (4)
interrogations, 235
intonation, 227
J, 137
[k], how pronounced, 3, 155
1, disappearance of, 186 (3) ;
final, no, 115; often
dropped in il, Us, etc., 116
[1], how pronounced, in ; voiced,
106; doubled, 172 (4) ; often
doubled after a pronoun, 1 1 6 w.
1-mouille, 107, 113
labio-dental consonants, 107
laterals, 107
Latin words, 86 (4), 129 (2), 15
(2), 104 (4) 2 (b), 157, 167,
172 (4)
Latin accent, 194
Law of Three Consonants, 176;
exceptions to, 178
le, under stress, 104 (i)
lengthening consonants, 104 (4)
liaison, 204- ; o in, 50 n. ; in
nasal vowels, loi ; none
before numerals, 170 ; after
numerals, 170
lip-rounding, 4, 9 (3), 45, 55, 59,
63, 69
low tone, 246 (3)
m, sounded falsely, 90 ; followed
by vowel, 86 (i) ; final,
sounded, 86 (4)
[m], 7 ; voiced, 106 ; doubled,
172 (4) ; initial, 224;^ (3)
INDEX OF SUBJECTS
mm, 86 (2)
mn, 86 (3)
mats, in liaison, 209 (i) n.
middle tone, 246 (2)
middle e, 28 ; a, 43 (4) ; o, 51 (3)
mixed vowel-sounds, 4, 58, 69
tnomeniandes, 179 (i)
mouth, in vowel sounds, 9 (2)
n, followed by vowel, 86 (i) ; final,
sounded, 86 (4)
n-mouille, 141.
[n], sounded falsely, before d or t,
90 (1) ; before p or b, 90 (2) ;
before c or g, 90 (3)
[n], doubled, 172 (4) ; voiced, 106
[p], 3 J how pronounced, 141
nasal consonants, 107, 139- ;
table of, 142
nasal vowels, table of, 4 ; how
pronounced, 83, 87 ; in
ordinary spelling, 84 ; dura-
tion, 85 ; liaison in, loi
naturally long vowels, 40, 47, 64
-ng, English, 87, 89, 141
nn, 86 (2)
non-accentuation, 200
numerals, 168- ; no liaison or
elision before, 170, 189 (3),
209 (5), 211 (4)
o, close, 44- ; open, 48 ; close and
open, 44 ; final, 47 (i) ;
before [z], 47 (4), 49 n. ;
in closed syllables, 50 (i) ;
in open non-final syllables,
50 (2) ; in derivatives, 51 (i) ;
middle, 51 (3) ; unstressed,
51-
M, 51 (3)
6, 44, 47 (2)
[o], how pronounced, 4, 8, 44, 45 ;
long and short, 47 ; in
ordinary spelling, 47 ; in
liaison, 50 n. ; duration,
104 (3)
[5], how pronounced, 4, 93 ; dis-
tinguished from [a], 94 ; in
ordinary spelling, 95
[o], how pronounced, 4, 8, 44, 48 ;
duration, 49, 104 {4) ; in
237
ordinary spelling, 50 ; un-
stressed, 49
[0], how pronounced, 4, 63 ; long
and short ; in ordinary
spelling, 64
oa, 44
oe, Greek, 41 (4) ; French, 66
[ce], how pronounced, 4, 65 ; in
ordinary spelling, 66, 67 ; un-
stressed, 66 ; duration, 66,
104 (4)
[oe], how pronounced, 4, 100
06, how pronounced, 38 (7)
06, how pronounced, 40, 72 (3)
oe, how pronounced, 38 (7), 72
(3)
oeu, 66
oi, 37 ; how pronounced, 38 (4),
72 (2) ; after r, 38 (4), 40
ol, 37 ; how pronounced, 38 (4),
40, 72 (2) ; after r, 41 (7)
oie, 37 ; how pronounced, 38 (6),
41 (7), 72 (2) ; after r, 41 (7)
onze, in liaison, 211 (4) n.
00, 50 (5)
open vowels, 9 (2)
opposition, emphasis in, 220
orthography, French, i
ou, 53 ; followed by vowel, 72 (i)
ou, 53
oue, 37 ; how pronounced, 38
(7)
oui, no liaison before, 211 (5)
ouaie, in liaison, 211 (5)
oy, 37 ; how pronounced, 38 (5),
72 (2) ; after r, 41 (6)
w, Greek, 50 (i) n.
p, medial, 146 ; final, 147 ; in
liaison, 205
[p], how pronounced, 145; doubled,
173 (4)
palate, 87
palatal consonants, 107
palatal r, 118, 120 w.
Parisian r, 119, 120 w.
pataques, 208 n.
parentheses, 233
ph, 124
pharynx, 87 (2)
phonetic system, i- ; alphabet,
phrases in liaison, 209 (6)
^
23S MANUAL OF FRENCH PRONUNCIATION
plural s in liaison, 211 (i) ; in
compound nouns, 211 (2)
prefix re-, 184
prepalatal consonants, 107
pretonic syllables, 104 (4) 3
poetry, see verse
q, for [k], 156
qu, 157
quandy in liaison, 209 (i) n.
quantity, 103-
quatre, in liaison, 209 (5) n.
r, disappearance of, 186 (3) ; final,
no, 123 ; followed by mute
consonant, in liaison, 211 (i)
r-grasseye, 119
[r], how pronounced, 117, 118 ;
voiced, 106 ; lengthening
consonant, 104 (4) ; doubled,
172 (4)
[r], how pronounced, 119, 120 n.
rang, in liaison, 209 (4) n.
respect, in liaison, 209 (4) n.
rhythm, 248 ; discordant, 252
rhythmic accent, see accent
roi, roi, 41 (5)
roie, 41 (7)
root-words, 104 (4) 3
rounding of lips, 4, 9 (3), 45, 55,
59, 63, 69
roy, 41 (6)
[-rwa-], 104 (3)
s, before b or m, 224 (2) ; after 1,
224 (4) ; final mute, in
liaison, 209 (4) n.
[s], 3 ; how pronounced, 127 ;
final, 128 ; final, sounded,
129 ; final, in liaison, 205 ;
medial, 128 ; between two
vowels, 132 ; before another
consonant, 128 (3) ; followed
by e mute, 185 ; plural, in
liaison, 211 (i) (2) ; doubled,
172 (4)
[J], 3 ; how pronounced, 134, 135
sang, in liaison, 209 (4) n.
sc, 130
Scotch sounds, 4, 23, 34, 36, 39,
44
semi-consonants, 3, 71 ; [w], 72 ;
[j], 73; M, 80
seven-vowel rule, 16, 24, 37, 49,
56, 61, 66, 104 (4)
slack vowels, 11
soft consonants, 106
sound-shading, 245
sounds, duration of, 103-
spelling, ordinary, i
stem, 104 (4) 3
stress, influence on length of
vowel, 104 (3), 104 (4) ; le
under, 104 (i) ; difference of
in derivatives, 43 (2)
stress, emphatic, [e] and [9] in,
104 (i), 214- ; see under
accent
stress-group, 16, 104 (4) 3, 196 ;
liaison in, 207 ; in elevated
speech, 203
stress-groups, in eurhythmy, 248
syllables, division of, 5, in verse,
75, 86 n. ; open 5, 104 (4) 3 ;
closed, 5, pronunciation in,
104 (4) 3 ; pretonic, 104 (4)
3 ; e mute in initial, 1 79
t, medial, 152 ; final, 151
[t], how pronounced, 60, 148, 149;
English, 149; doubled, 172
(4)
tenir, -ien in, 99
tenseness in vowel-sounds, 11, 19,
35, 46
th, 148, 152
ti-, before vowel, 150
tone, high or low, 246
tongue, in vowel sounds, 11, 19,
. 35, 46
tonic words, 236
tonic accent, see accent
toujours, in liaison, 211 (i)
tout, in liaison, 209 n.
triangle of vowels, 10, 58, 102
trills, 107
u, 62, 80 n.
[u], how pronounced, 4, 8, 53,
54 ; duration, 56, 104 (4) ;
unstressed, 56
[u] (open), 53
ue, 66
INDEX OF SUBJECTS
239
uhlan, no liaison before, 211 (5)
un (the numeral) in liaison, 211
(4)
unfamiliar words, 86 (4)
uniformity in vowel sounds, 12
unstressed syllables, 16 n.
unstressed vowels, 104 (3), 104
(4) 3
unvoiced consonants, 106
upward part of sentences, 231
uvula, 87
uvular consonants, 107
uvular [r], 119, 120
[v], lengthening consonant, 104
(4) ; how pronounced, 126
velar consonants, 107
velours, 208 n.
velum, 87
venir, -ien in, 99
verse, rhythm in, 254 ; [j] and
[i] in, 75 ; licence in rhymes,
104 (a) n. ', e mute in, 191 ;
liaison in, 212
vocalic assimilation, 29, 51 (2), 68
voiced consonants, 106
vowel, indeterminate, 69-
vowel-sounds, French, 4
vowels, defined, 7 ; fundamental,
7- ; front, 9 (i), 10 ; back,
9 (i), 10 ; close, 9 (2) ; open,
9 (2) ; tense and slack, 11 ;
clearness of, in French, 13 ;
the naturally long, 40, 47,
64, 104 (3) ; half-long, 40 ;
mixed, 4, 58 ; triangle of,
10, 58, 102 ; when final,
104 (2) ; circumflexed, 104
(4) 2 (a)
vowels, nasal, how pronounced,
83, 87 ; in ordinary spelling,
84 ; duration, 85
w, in English words, 72 {5) ; in
German and Flemish words,
126
[w], 3, 7, 72
words, tonic, 236 ; compound,
132 ; in combination, 174 ;
fixed, 1 88
X, for [s], 130; for [k], 156; for
[ks], 158; final, 159; in
liaison, 205 ; initial, 158 n.
y, between two vowels, 77, 81;
initial, in liaison, 211 (5)
y (the pronoun), 79
[y], how pronounced, 4, 58, 59 ;
after [r], [s], and [t], 60 ;
unstressed, 60 ; duration, 61,
104 (4) ; in ordinary spelling,
62
M> 3> 7 ' how pronounced, 80
yod, 73-
[z], 3 ; lengthening consonant,
104 (4) ; how pronounced,
131 ; final, 133
[3], 3 ; lengthening consonant,
104 {4) ; how pronounced,
137
zh, 137, 160
BIBLIOGRAPHY
FOR THE FURTHER STUDY OF THE
SOUNDS OF FRENCH i
{As revised for this Manual by M. Paul Passy, D.-is-L., Paris)
(Works are here placed in alphabetical order. Those marked
with an asterisk (*) employ the alphabet of the International
Phonetic Association.")
I. GENERAL PHONETICS
♦Althaus (L. H.) : The Means of Training in Phonetics available
for Modern Language Teachers. L' Association Phonetique
Internationale.
♦Beyer (F.) : Franzosische Phonetik. Schulze, Cothen.
BouRCiEZ (E.) : Precis historique de phonetique frangaise. 1907,
Ellis (A.) : A Plea for Phonetic Spelling. 1848.
FELINE : Methode pour apprendre d lire. 1854.
♦Jespersen (O.) : Lehrbuch der Phonetik. Teubner, Leipzig,
1904.
„ Phonetische Grundfragen. Teubner, Leipzig,
1905.
♦Jones (D.) : Examination Papers in Phonetics. Clarendon
Press, Oxford, 191 9.
♦Klinghardt (H.) : Franzosische IntonationsUbungen. Schulze,
Cothen.
♦Nicholson (G. G.) : Introduction to French Phonetics. Mac-
millan, London.
♦Noel-Armfield (G.) : General Phonetics. Heffer, Cambridge,
1915.
* Works out of print may be consulted in libraries, or may sometimes
be obtained second-hand.
2 For works published by the International Phonetic Association,
application may be made to the English Secretary, Mr D. Jones,
University College, London, W.C.i.
240
BIBLIOGRAPHY 241
♦Palmer (H. E.) : What is Phonetics ? L' Association Phon6-
tique Internationale, 1915.
♦Partington (V.) : The Transition from Phonetic to Ordinary
Spelling. L' Association Phonetique Internationale.
»^Passy (P.) : Petite Phonetique comparee des principales langues
europ^ennes. Teubner, Leipzig, 191 2.
•^ ,, La Phonetique appliqtcee d I'enseignement de la
lecture. L' Association Phonetique Interna-
tionale, 191 6.
„ Le Phondtisme au congrh de Stockholm. 1887.
,, L' Instruction primaire aux ^tats-Unis. 1885.
♦Perret (W.) : Some Questions of Phonetic Theory. Heffer,
Cambridge, 1917-
♦RippMANN (W.) : Elements of Phonetics, English, French, and
German, translated and adapted from Victor's " Kleine
Phonetik." Dent, London, 1899.
RoussELOT (P.) : Principes de phonetique experimental .
Welter, Paris, 1902.
♦ScHOLLE and Smith : Elementary Phonetics. Blackie, London,
1908.
Scripture (E.) : Elements of Experimental Phonetics. Arnold,
London, 1904 ; Scribner, New York.
♦SoAMES (L.) : Introduction to Phonetics. Sonnenschein, London,
1899.
Sweet (H.) : Primer of Phonetics. Clarendon Press, Oxford,
1906.
♦ViETOR (W.) : Elemente der Phonetik. Reisland, Leipzig, 1904.
* ,, Kleine Phonetik. Reisland, Leipzig, 1903.
*L' Association Phonetique Internationale : L'^criture
phonetique Internationale, and various other pamphlets.
2. FRENCH PRONUNCIATION AND DICTION
♦Baker (A.) : Historical French Grammar. Dent, London, 1900.
\JDomergue (U.) : La Prononciation frangaise ddterminee par des
signes invaridbles. Paris, 1797.
x^^UMViLLE (B.) : Elements of French Pronunciation and Diction.
Dent, London, 1904.
Dupuis (Mme) : Traite de prononciation ou nouvelle prosodie
frangaise. 1836.
M2 MANUAL OF FRENCH PRONUNCIATION
' DjtrpoNT- Vernon : Principes de diction.
V „ L'Art de Men dire
^ tions. 1888.
1882.
principes et applica-
xii^ sidcle. 1845.
y^
Genin (F.) : Variations depuis
GoEMANS (L.) et Gri^goire (A.) : Petit Traite de prononciation
frangaise. E. Champion, Paris, 1919 ; Benard, Liege.
Grace (S. W.) : Primer of French Pronunciation. Methuen,
London, 1920.
Grammont (M.) : Traite pratique de prononciation frangaise.
Delagrave, Paris, 1914.
Martinon (Ph.) : Comment on prononce le frangais. Larousse,
Paris, 1913.
Nyrop (K.) : Grammaire historique de la langue frangaise.
Copenhagen, 1904.
* „ Manuel phonetique du frangais parU. Traduit et
remanie par E. Philipot. Picard et Fils, Paris,
/ 1 91 4 ; G. E. Stetchert, New York.
/ Paris (G.) : J^tudes sur la role de V accent latin. 1862.
*Passy(P.) : Les Sons du frangais. Firmin-Didot, Paris, 1906;
Didier, Paris, 191 2.
* ,, The Sounds of the French Language. A Transla-
tion of the above, by D. L. Savory and
Daniel Jones. Clarendon Press, Oxford,
2nd Edition, 191 3.
* ,, Abrege de prononciation frangaise {phonetique et
orthoepie). Reisland, Leipzig, 191 3.
* ,, Changements phonetique s. Paris, 189 1.
♦Quiehl (K.) : Franzosische Aussprache. Elwert, Marburg, 1899.
♦Richards (S. A.) : French Speech and Spelling, a First Guide
to French Pronunciation. Dent, London, 1907.
Rosset : Les Origines de la prononciation moderne. 191 1.
RouDET (L.) : La Desaccentuation et le deplacement de Vaccent
dans le frangais moderne. Dans la Revue de philologie
frangaise, 1907.
RoussELOT (P.) et Laclotte (F.)
\y
*Saillens (£.) and Holme (E. R.
Pronunciation. Blackie, London, 1909.
Premiers elements de pro-
nonciation frangaise. H.
Welter, Paris.
Precis de prononciation
frangaise. H. Welter,
Paris, 1902.
First Principles of French
BIBLIOGRAPHY 243
Scripture (E.) : Speech Curves. Carnegie Institution, Washing-
ton, 1906.
^.-'^HURQT (F. C.) : La Prononciation frangaise depuis le commence-
ment du xvi« sidcle. Imprimerie Nationale, 1881.
Lesaint (M. a.) : Traiti complet de la prononciation frangaise.
Revu et compl^t^ par Chr. Vogel, 1890.
3. PHONETIC READERS
♦Jones (D.) : Cent podsies enfantines. Teubner, Leipzig.
* „ Intonation Curves. Teubner, Leipzig.
KoscHWiTZ (E.) : Les Parlers parisiens. Anthologie phond-
tique. Revue par Franz, 191 1.
*Noel-Armfield (G.) et Brandin (L. M.) : Un peu de rire.
Heflfer, Cambridge.
♦Partington (V.) : French Songs and Speech. Marshall, London.
♦Passy (P.) : Premier livre de lecture. Firmin-Didot, Paris, 1884;
Didier, Paris, 191 7.
* „ Deuxidme livre de lecture. Idem.
Versions populaires du Nouveau Testament. Paris,
Soci6t6 des Trait6s, 1893-96.
Histoires pour enfants. Idem.
Le frangais parU. Reisland, Leipzig, 1914.
Choix de lectures. Schulze, Cothen, 1905.
Lectures varices. Soci6t6 des Trait^s, Paris, 1910.
A French Phonetic Reader. University of London
Press, 1914.
Lectures frangaises phondtiques. Heffer, Cam-
bridge, 1918.
Conversations frangaises. University of London
Press, 1920.
♦Passy (J.) et Rambaud (A.) : Chrestomathie frangaise. Leipzig,
1908.
*
ICHARDS (S. A.) : French Phonetic Reader. Dent, London, 1912.
♦RossET (Th.) : Exercices pratiques de prononciation et de diction.
Grenoble.
♦Savory (D. L.) : Premiere annee de frangais. Black, London.
♦ „ French Oral Teaching. Rivingtons, London.
♦Spiers (V.) : Senior French Reciter. Simpkin, Marshall & Co.,
London, 1909.
V ♦Thudichum (G.) : Manuel de diction. Kiindig, Geneva.
244 MANUAL OF FRENCH PRONUNCIATION
*Le MAfTRE PHON]feTiQUE, bi-monthly Journal of the Inter-
national Phonetic Association.
4. DICTIONARIES
Hatzfeld et Darmesteter : Diciionnaire general de la langue
jrangaise. Delagrave, Paris.
♦MiCHAELis et Passy : Dictionnaire phonetique frangais. Meyer,
Hanover, 1897.
♦Pierce (R. M.) : International French and English Dictionary,
Jack, London, 1904.
♦Rangel and Nielsen : Fransk-Dansk Ordbog. Bojesen,
Copenhagen, 1903.
♦Baker (E. A.) : French and English Dictionary. Cassell,
London and New York, 1920.
♦Pfohl (E.) : Neues Wdrterbuch der franzdsischen und deutschen
Sprache. Brockhaus, Leipzig.
♦Edgren-Burnett : French and English Word-book. Heine-
mann, London, 1902.
5. PHONETIC CHARTS (Paper or Linen)
♦Jones (D.) : Les Sons du frangais. Cambridge.
♦ „ The Organs of Speech. Cambridge.
♦Jones (D.) and Rausch (F.) : A Set of Sound Charts (9),
showing Tongue and Lip Positions. Dent, London.
♦RippMAN (W.) : Les Sons du frangais. Three Charts, size
30x30 inches. Small reproductions in packets of 30.
Dent, London.
♦Scholle and Smith: Coloured Wall- Charts of French Sounds.
Blackie, London.
♦ViETOR (W.) : Tableau des sons frangais. Elwert, Marburg.
Printed in Great Britain by Turttbull &fi Spears^ Edinburgh
PC Jack, Jaiaes William
2137 Manuel of French
J3 pronunciation and diction
cop, 2
PLEASE DO NOT REMOVE
SUPS FROM THIS POCKET
UNIVERSITY OF TORONTO
LIBRARY
1
unfrnmH
HI
3
i . .
^^H
*"
1
.: .: :-v::;.:;;-;;:J'
HlfflnBf
''
S^^i^?iivi!*tr
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Books to Borrow
Open Library
TV News
Understanding 9/11