PosterPDF Available

Gender and acoustic variation: cases of vowels in Arabic Libyan of Tripoli

Authors:

Abstract

This study is about the differences in articulations between men and women in Libyan Arabic from Tripoli and the social power structure of these differences. The corpus used is a list of words with consonants having strong distinctive values to a social point of view between men and women, i.e, pharyngealized consonants / s ˁ, t ˁ, d ˁ/ and non-pharyngealized consonants /s, t, d/. Gender differences will be treated here through values of frequency of the three first formants [F1, F2, F3] of the vowels /i, u, a/ which are linked to the pharyngealized and non-pharyngealized consonants. The results show significant between men and women and these differences are caused by social distinctions.
Sociolinguistics Symposium 19: Freie Universität Berlin | August 21-24, 2012
Gender and acoustic variation: cases of vowels in Arabic Libyan of Tripoli
FATHI SALAM & MOHAMED EMBARKI
LLC-ELLIAD, Université de Franche-Comté, Besançon
30 rue Mégevand 25030 BESANÇON Cedex
fst2005@hotmail.com
ABSTRACT
This study is about the differences in articulations between
men and women in Libyan Arabic from Tripoli and the
social power structure of these differences. The corpus
used is a list of words with consonants having strong
distinctive values to a social point of view between men
and women, i.e, pharyngealized consonants / s ˁ, t ˁ, d ˁ/
and non-pharyngealized consonants /s, t, d/. Gender
differences will be treated here through values of fre-
quency of the three first formants [F1, F2, F3] of the
vowels /i, u, a/ which are linked to the pharyngealized and
non-pharyngealized consonants. The results show
significant between men and women and these differences
are caused by social distinctions.
1- Acoustic parameters of
pharyngealized and non-
pharyngealized consonants
F1 high in contact with pharyngeal-
ized consonants= due to the reduc-
tion of the pharynx cavity.
F2 low in contact with pharyngeal-
ized consonants= due to the exten-
sion of the pharynx cavity.
The literature shows that the
consonants adjacent to the pharyngealized consonants have
consistently higher values of F1 and lower values of F2,
compared to the same vowels before the non-pharyngealized
consonants.
pharyngealized non- consonants pharyngealized consonants 2- Gender and frequential values
Both gender have different vocal canal= it is on average 15%
shorter for a female adult compared to that of a male adult. As
a consequence= the average frequency of the formants is
15% higher for women compared to men. The studies show
that the vowels in front of the pharyngealized consonants and
non-pharyngealized are characterized by a higher value for
women than for men.
3- coarticulation
Speech is not made of isolated gestures but it is rather made
of sounds linked one to another; they are said to be
co-articulated (Embarki & Dodane, 2011). Lindblom (1963)
has tried to characterize the coarticulation between two
segments, the consonant (C) and the vowel (V), through a
linear regression known as locus equation. As for the Arabic,
the first study to have put into practice is Sussman’s (1993)
on the Cairo Arabic. Yeou (1997) has used the locus equation
to a bigger corpus in standard Arabic. Yeou shows that the
locus equation allows to clearly distinguish the pharyngealized
and non-pharyngealized consonants and the coarticulation
between C and V depends on the nature of the segments.
INTRODUCTION
HYPOTHESIS
The variation of mean value of F1 and F2 of /I, u a/ de-
pends on the consonantal context, notably if it is
pharyngealized or non- pharyngealized .
The slope of the locus equation will be a distinctive sign
between men and women.
The gender difference (men Vs women) has a direct im-
pact on how to create the syllable CV (C= consonant ; V=
vowel).
10 speakers have taken part in the experiment, 6 men and 4
women, aged 17 to 30. The corpus is made of a list of words with
or without meaning (logatome). The words are trisyllabic (C1V1
C2V2 C
3V3) C= /s sˁ t tˁ d dˁ/V= /i, u, a/. The vowel has been
measured in the beginning, in the middle and in the end.
Regarding the measurement of the locus equation: F2 has been
measured in the beginning (F2 onset) and in the middle (F2Mid).
All the words have been split up and manually labeled under
PRAAT.
METHODOLOGY
RESULTS
The values of the slope of the pharyngealized Vs the non-pharyngealized consonants is higher for women than for
men (this result is conform to the results of the study on the Kowaiti Arabic (Embarki and Ahmad, 2010).
An increase of the values of F1 and a decrease of F2 for the pharyngealized consonants. An
increase of the values of F1 and a decrease of F2 for the non- pharyngealized consonants. An
increase of the values of F3 for the pharyngealized consonants.
Values of F1 & F2 of vowels /i, u, a/ in the contact of pharyngealized and non-pharyngealized in Libyan Arabic
of Tripoli
As in literature, mean values of the vowels /i, u, a/ are higher for women than for men.
F1 F2 F3
PH N-Ph Ph N-Ph Ph N-Ph
wo-
men
i Moy 350 305 2211 2626 3174 3331
E.T 94 7 401 24 287 53
u Moy 358 347 1211 1228 3042 2936
E.T 64 65 265 332 238 309
a Moy 686 621 1486 1819 3247 3054
E.T 65 81 173 412 306 369
men
i Moy 327 294 1949 2299 2875 3148
E.T 43 20 287 61 176 106
u Moy 325 321 960 912 2868 2562
E.T 29 14 184 51 181 76
a Moy 500 478 1263 1647 2991 2844
E.T 33 26 78 74 125 165
F1 F2
PH N-Ph Ph N-Ph
i
Moy 331 304 2110 2377
E.T 70 44 305 237
u
Moy 323 322 961 1185
E.T 20 52 94 345
a Moy 567 478 1304 1647
E.T 113 26 213 74
Values of F1, F2 and F3 of vowels /i, u, a/ in the contact of pharyngealized and non-pharyngealized according to the
gender
The pharyngealized consonants are distinguished form non- pharyngealized
consonants in Arabic libyan of Tripoli, i.e. the slope values of the
pharyngealized consonants are flatter than those of their
non- pharyngealized. This result is conform to the results of the study on the
Arabic: Sussman et al. (1993); Yeou (1997), Embarki et al. (2011).
RESULTS OF LOCUS EQUATION
non- pharyngealized pharyngealized
V1 t d s tˁ dˁ sˁ
Inter-y 388 675 578 405 575 478
Slope 0,774 0,649 0,693 0,632 0,526 0,643
R2 0,842 0,629 0,715 0,699 0,488 0,789
Mean values of y intercepts (inter y), slopes and regression coefficients (R2) corresponding to data
form Arabic libyan of Tripoli. Averaged 10 speakers.
d
dˁs
sˁ
t
tˁ
350
400
450
500
550
600
650
700
0,5 0,5 5 0,6 0,6 5 0,7 0,7 5 0,8
Intery
Pente
Locus equation of [t] (left) and [tˁ] (right), Averaged 10 speakers.
Valeus of locus equation of pharyngealized consonants and non- pharyngealized
consonants in Arabic libyan of Tripoli. Averaged 10 speakers
non-pharyngealized pharyngealized
t d s tˁ dˁ sˁ
Women
Inter-y 324 728 450 509 921 537
slope 0,817 0,633 0,768 0,647 0,562 0,646
R2 0,902 0,560 0,715 0,723 0,281 0,831
Men
Inter-y 460 637 668 439 590 687
slope 0,727 0,662 0,633 0,576 0,502 0,604
R2 0,780 0,697 0,710 0,711 0,496 0,753
Locus equation of [t] (left) for women, (right), for men
Mean values of y intercepts (inter y), slopes and regression coefficients (R2) for /tˁ, dˁ, sˁ/ and /t, d, s/ according to the gender
The visible influence of the consonantal pharyngealized context Vs non- pharyngealized on the building of the vowel
within the syllable (CV). The locus equation, extended here to the study of gender, shows that the values of slope are
different. The results tend to show a differentiated tendency according to the gender to build the most minimal linguistic
units as the syllable.
CONCLUSION
Al-ani, Salman, (1970). Arabic phonology, Mouton. Al-Tamimi, Feda et Heselwood, Barry, (2011). Nasoendoscopic, videofluroscopic and acoustic study of plain and emphatic coronals in Jordanian Arabic, Zeki Hassan et Heselwood Barry (éds), instrumental studies in Arabic Pho-
netic, P. 156-191, John Benjamins publishing company. Embarki, Mohamed & Ammar, Ahmad, (2010). L’équation de locus comme mesure de distinction sociale de gender en arabe koweïtien, actes des XXVIIIèmes JEP, Mons, 25-28 mai. Embarki, Mohamed.Dodan Christelle
( 2011) La coarticulation des indices à la representation, L’Harmattan. Embarki, Mohamed., Ouni, Slim., Yeou, Mohamed., Guilleminot, Christian & Al-Maqtari, Sallal, (2011b). Acoustic and electromagnetic articulographic study of pharyngealisation: Coarticulatory effects as an
index of stylistic and regional variation in Arabic, Hassan Z, M et Heselwood, B (éds), instrumental studies in Arabic Phonetics, John Benjamins Publishing Company. Ghazeli, Salem, (1981). La coarticulation de l'emphase en arabe, Arabica, T. 28, Fasc. 2/3, Numéro Spécial Double :
Études de Linguistique Arabe (Jun. -Sep), P. 251-277. Hassan, Majeed et Esling, John, (2011). Investigating the emphatic feature in Iraqi Arabic: Acoustic and articulator evidence of coarticulatoire, Hassan Z, M et Heselwood, B (éds), instrumental studies in Arabic Phonetics, John
Benjamins Publishing Company. Kahn, Margaret, (1975). Arabic Emphatics: the evidence for cultural determinants of phonetic Sex-typing, phonetica, n°31, P. 38-51. Khattab, Ghada., Al-Tamimi, Feda et Heselwood, Barry, (2006). Acoustic and Auditory differences in the /t/- /T/ Op-
position in Male and Female Speakers of Jordanian Arabic, Boudelaa Sami (Ed.), Perspectives on Arabic Linguistics XVI: Papers from the sixteenth annual symposium on Arabic linguistics PP 131-160 Cambridge, UK: John Benjamins. Lindblom, Bjorn, (1963). Spectrographic study
of vowel reduction, journal society American, vol. 11, P. 1773-1781. Jongman, Allard., Herd, Wendy., Al-Masri Mohammad., Sereno Joan, et Combest, Sonja (2011). Acoustics and perception of emphasis in urban Jordanian Arabic, journal of phonetics, vol. 39, P. 85-95. Sussman,
Harvey., Hoemeke Kathryn et Ahmed Farhan, (1993). A cross-linguistic investigation of locus equation as a phonetics descriptor for placed of articulation, Acoustical society of America, vol 94, n° 3, P. 1256-1268. Yeou, Mohamed, (1997). Locus equation and the degree of coarticu-
lation of Arabic consonants, phonetica, n° 54, P. 187-202.
0
500
1000
1500
2000
2500
3000
iuaiua
Women Men
F1ph F1Nph F2ph F2Nph
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.