Scholarly article on topic 'Glottal Stop in Hiatus: An Acoustic Investigation in Persian'

Glottal Stop in Hiatus: An Acoustic Investigation in Persian Academic research paper on "Computer and information sciences"

CC BY-NC-ND
0
0
Share paper
Keywords
{"Glottal stop" / hiatus / "acoustic analysis" / "formant transition ;"}

Abstract of research paper on Computer and information sciences, author of scientific article — Samaneh Yazarlou

Abstract One of the strategies to resolve hiatus in Persian is glottal stop (GS) insertion between two adjacent vowels (Sadeghi, 2001: 27, 28) [1]. This paper investigates the presence of GS in hiatus within and across morpheme boundaries in isolated words (684 tokens) as well as in words within sentences (276 tokens). Four speakers of standard Persian (two females and two males) between the ages of 31 and 37 participated in this study. Acoustic analysis of the data by Praat [2] showed three acoustic correlates of GS in intervocalic environments; in addition to a complete glottal closure and release (GC) and creaky voice (CV), completely irregular vibration of vocal folds (CIVV) was also observed. CV was characterized by a low fundamental frequency, a semi-periodic waveform and a decreasing intensity level, while no fundamental frequency, an aperiodic waveform and a descending intensity level characterized CIVV. The majority of GS presence was observed where a derivational prefix and a base had conjoined (11.9% GC, 27.38% CV and 55.95% CIVV). The significant outcome of data analysis in other morpheme boundaries and in simple words showed that GS did not exist/insert in hiatus in a great number of data; in 77.30% of isolated tokens and in 79.71% of in-the-context tokens two neighboring vowels in sequence were observed showing formant transition. This considerable number of data demonstrated that in standard formal Persian it is more probable to have two adjacent vowels in simple words and in morphological boundaries (except in derivational prefix + base environment) than to insert a GS.

Academic research paper on topic "Glottal Stop in Hiatus: An Acoustic Investigation in Persian"

Available online at www.sciencedirect.com

ScienceDirect

Procedia - Social and Behavioral Sciences 136 (2014) 12 - 20

LINELT 2013

Glottal Stop in Hiatus: An Acoustic Investigation in Persian

Samaneh Yazarlou a

aDepartment of Linguistics, Allameh Tabataba'i University, South Allameh Tabataba'i St., Modiriyat Bridge, Shahid Chamran Highway, Tehran, 1997967556,

Abstract

One of the strategies to resolve hiatus in Persian is glottal stop (GS) insertion between two adjacent vowels (Sadeghi, 2001: 27, 28) [1]. This paper investigates the presence of GS in hiatus within and across morpheme boundaries in isolated words (684 tokens) as well as in words within sentences (276 tokens). Four speakers of standard Persian (two females and two males) between the ages of 31 and 37 participated in this study. Acoustic analysis of the data by Praat [2] showed three acoustic correlates of GS in intervocalic environments; in addition to a complete glottal closure and release (GC) and creaky voice (CV), completely irregular vibration of vocal folds (CIVV) was also observed. CV was characterized by a low fundamental frequency, a semi-periodic waveform and a decreasing intensity level, while no fundamental frequency, an aperiodic waveform and a descending intensity level characterized CIVV. The majority of GS presence was observed where a derivational prefix and a base had conjoined (11.9% GC, 27.38% CV and 55.95% CIVV). The significant outcome of data analysis in other morpheme boundaries and in simple words showed that GS did not exist/insert in hiatus in a great number of data; in 77.30% of isolated tokens and in 79.71% of in-the-context tokens two neighboring vowels in sequence were observed showing formant transition. This considerable number of data demonstrated that in standard formal Persian it is more probable to have two adjacent vowels in simple words and in morphological boundaries (except in derivational prefix + base environment) than to insert a GS. © 2014 The Authors.PublishedbyElsevierLtd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of the Organizing Committee of LINELT 2013. Keywords: Glottal stop, hiatus, acoustic analysis, formant transition;

1. Nomenclature

2. DS derivational suffix v verb

3. DP derivational prefix C clitic

4. N noun Fo fundamental frequency

5. Adj. adjective V vowel

6. SW original simple word(s) F2 second formant

Corresponding author: Samaneh Yazarlou / Tel.: +98-912-620-4454 E-mail address: samanehyazarlou@yahoo.com

1877-0428 © 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of the Organizing Committee of LINELT 2013. doi:10.1016/j.sbspro.2014.05.279

7. IP inflectional prefix RB release burst

8. All transcriptions of examples in the article are based on their reference style.

1. Introduction

GS in hiatus has long been the center of attention for many grammarians and linguists of Persian. Like many languages of the world which use strategies such as heterosyllabification, diphthong formation, epenthesis, vowel elision, glide formation and coalescence to resolve hiatus (Casali, 1996: 1) [3], Persian uses strategies such as deletion of the intervocalic consonant and changing the second vowel into a glide, e.g. miya'i 'you are coming' > miyay, usually in colloquial or fast speech (Lazard, 1957 [2005]: 21) [4], or a consonant insertion in intervocalic environments, e.g. [Q], i.e. hamze a, or [-y-] appearance in [-e + -iDS] environment in mohavere-'/y-i 'conversational' to deal with hiatus (Sadeghi, 2001: 8) [1]. Although studies about GS in Persian are well documented, less is known about its acoustic characteristics in intervocalic positions, in different morphological boundaries and in original simple words. This issue is the central focus of this paper.

Nye (1954: 10) [5] and Lazard (1957 [2005]: 39) [4] posit that GS between two vowels exists in careful speech, e.g. /sa?aPt/ 'hour'; mo'^llem 'teacher'. Similarly, Stagi et al. (2010: 47, 48) [6], in an acoustic investigation, illustrate that GS presence between two vowels is restricted to careful speech (except in low frequency or formal words in which GS exists even in fast speech).

Nye (1954: 31, 32) [5] believes that GS in intervocalic environments might be omitted with one of its accompanying vowels in colloquial speech, e.g. /sa?aPt/ > /sat/ [sa:t] 'hour'. He (1954: 17) [5] also indicates the possibility of sequences of vowels in cross morpheme boundaries in Tehrani dialect of modern Persian which always occurs by crossing a syllable boundary, e.g. /bi-/ imperative prefix + /-a-/verb root 'come' + /-id/ second person plural suffix >

/biP=id/ 'Come'. Similarly, Lazard (1957 [2005]: 39, 42) [4] suggests GS deletion in colloquial speech, e.g. mo^llem 'teacher' and states the possibility of hiatus occurrence in these instances: [i + a] > zendeEi-am 'my life', [e + a] > xane-am 'my house', [i + e] > zendeEi-esan 'their life'. Yarmohammadi (1964: 13) [7] also addresses the same idea in informal speech, e.g. /so [sa P^>J1] 'clock/hour'. Haghshenas (2003: 270) [8] postulates that

all phonetic representations of GS between two vowels 'might' be omitted in fast speech. An acoustic investigation by Stagi et al. (2010: 39, 47, 48) [6] also illustrates this idea. Lambton (1971: xv) [9] believes that hamze only occurs in the medial position in orthography, not the transcription, e.g. pa/z 'autumn', d=n=i 'a wise person'.

Windfuhr (1979: 139-140) [11] posits that the prescriptive orthography has led to the assumption that the correct reading/pronunciation of the orthographic symbol hamze in hiatus should accompany GS insertion. Similarly, Jahani (2005: 81) [12] suggests that GS pronunciation in Persian words, such as p=& ^ iz 'autumn' and gol-e xosbu»i 'a pleasant smelling flower' is artificial and is based on the written language. In Jahani's view, the hiatus-filler employed in these words in the spoken language is glide y, e.g. p= ltyiz and gol-e xosbuyi.

Haghshenas (1978:34, 35) [10] attests that hamze 'could' be pronounced by glottal vibration between two vowels in words like [mo"ayene] 'examination' and in morpheme boundaries, e.g. [bi-]DP+[ Pab]N>[bi"ab]AdJ. 'waterless'. He believes that this glottal vibration is represented by short-time irregular amplitude, which is different from that of vowels. An alternative view is that of Parmun (2002:538, 550) [13] who postulates that GS does not exist in standard colloquial Persian as a mediator consonant, except in some artificial styles. Instead, he claims CV presence in some intervocalic positions in morphological boundaries.

The following study investigates the possibility of GS presence in intervocalic environments, in both morpheme

a In Persian orthography different phonetic representations of GS is usually referred to as hamze (Haghshenas, 1978: 23) [10].

boundaries and in original simple words, and also determines the acoustic characteristics of its allophones (in case of its presence). According to Persian syllable structure (cV, cVC, cVCC, CV, CVC, CVCC) which does not tolerate hiatus (Haghshenas, 1978: 31, 34, 54, 55) [10], it is hypothesized that GS exists/inserts between two adjacent vowels in morpheme boundaries and in original simple words.

2. Methods

The main objectives of this article were to investigate the possibility of GS presence in intervocalic positions, in both morpheme boundaries and in original simple words and to determine its allophones' acoustic characteristics in standard formal Persian.

2.1. Speakers

Four speakers of standard Persian (two females and two males), between the ages of 31-37, participated in this research. The females spoke English as a second language and the males spoke Turkish and English as their second and third languages respectively. In the presentation of the results the female speakers are referred to as FS1 and FS2 and the male speakers as MS1 and MS2.

2.2. Material

To study the acoustic characteristics of GS in intervocalic positions, two groups of data were selected: 1) 57 out-of-context words (OCW), and 2) 23 in-the-context words (ICW) (in 17 sentences). Intervocalic positions were in morpheme boundaries and in original simple words (Table 1). Two groups of data were chosen in the study to prevent ambiguity in the results, due to different orthographies of some words, like [p=jiz] / [p=»iz] 'autumn', and [d^n=ji] / [d^n=;»/J 'wisdom' in Persian; OCW were written with [j] and ICW with hamze in these cases but IP + v group in OCW had the same orthography as ICW. Each speaker read the written words and sentences three times. In total, [57*3*4] out-of-context tokens (OCT) and [23*3*4] in-the-context tokens (ICT) were investigated (figures in brackets show the number of words, repetition times and the number of speakers respectively).

2.3. Instruments/ Procedures

Subjects were recorded using a Hewlitt Packard laptop (model: ProBook 4520s), a Philips headset (model: SHM3300U/10) with noise cancelling microphone and Praat software (version 5132). The required data was recorded directly on Praat software with 22050 Hz sample rate frequency and the same software was used for all display, playback and measurement procedures.

Table 1. OCW, ICW (underlined) & the kind of morphemes in intervocalic positions

SW DP + base base + DS IP + v base + C base + C

[ >>=jin] [b=»^beru] [z=»u] [mi»istad] [»=rezuji]

[»=»in] 'reputable' 'in labor' 'she/he was standing' 'a rug' [ »=rezu»i]

'ritual' 'a wish'

[p4'z] [n=»omid] [d=n=ji] fmi»enO=radJ [G^ije cerm=n] [tarsu»e]

[p=»iz] 'disappointed' [d=n=»i] 'she/he assumes' 'Kerman rug' 'she/he's coward'

'autumn' 'wisdom'

[b=»adab] [bin=)i] [mi»and=xt] [C^ije<=n] [xodroji]

'polite' [bin=»i] 'she/he was throwing' [C^li»e<=n] fxodro»i]

'eyesight' 'their rug' 'a car'

[b=»e<tij^G] [r^dijoji] [mi»oft^d] [C^ijac] [r=hroje b4=]

'enthusiastic' [r^diio»i] [mi»oft^d] [r=hro»e b^=]

'radio-' 'it, she/he was falling' 'her/his rug' 'the upper corridor'

[bi»adab] [xo<c<uji] [mi»^mad] [sini-jo-fend /^n] [xodroje<=n]

'impolite' fxo<c<u»i] [mi»^mad] 'tray and cup' fxodro»e<^n]

[bi^im^n] 'faithless'

'dry cleaner's'

[labuji] [labu»il ■ 'a person who sells cooked beet'

[derje»u] 'crybaby'

[panbe»i] 'made of cotton'

[Gahve»i] 'brown'

'it, she/he was coming'

[be»ist] 'Stand up'

fbijenO=rad] 'she/he might assume'

[bi»afcan] 'Throw'

[bi»oft^d] 'it, she/he fell'

[r^h bi»oft] [r=h bi»oft] 'Start moving'

[ bijA/ar] 'Bring'

[na»ist^d] 'she/he didn't stand'

fnajenO=rad] 'she/he might not assume'

[najoft^d]

'it, she/he didn't fall'

[nal=JiJ=] [naj=ji»=]

'Don't come'

[x^ne»i] 'a house'

[x^neje<=n]

'their house'

[x^ne»a<] 'her/his house'

[x^ne-»o-k=<^ne] 'house and shelter'

fbipo <tv=ne»im] 'we're with no support'

fbipo <tv=ne»am] 'I'm with no support'

fbipo <tv=ne»id] 'youpl're with no support'

'their car'

[xodroja<] [xodro»a<l 'her/his car'

[m^nto-»o-p^lto] 'coat and overcoat'

[sed^'i] [sed=»i] 'a voice'

[zib=-»o-ze<t] 'beautiful and ugly'

fzib4imj [zib=»im] 'we're beautiful'

[zib^jam] fzib=»am] 'I'm beautiful'

[zib^jid] [zib=»id] 'youpl're beautiful'

• I. Kalbassi (1992 [2008]: 117) [14] 2.4. Measurements

Waveforms and spectrograms were used to acoustically investigate the phonetic representation(s) of GS in intervocalic positions. Due to different acoustic patterns of GS in intervocalic environments, their identification methods will be explained separately but vowel-to-vowel transition (VVT) will also be discussed in this section since it was observed in a great number of data.

2.4.1. Glottal closure

Complete glottal closure (GC) is represented by no waveform (Ladefoged & Maddieson, 1997: 76-77) [15]. It appeared in the spectrogram by no pitch line/no fundamental frequency and a sharply dropped intensity line. The reason that there was no F0 in GC measurements was that F0 is the acoustic correlate of vocal folds vibration (Modarresi Ghavami, 2012: 111) [16]. Since complete glottal closure accompanies no vibration, no F0 could be calculated in its measurements. F0 can be obtained from the software when glottal vibrations are regular, like vowels, or slightly irregular, like CV. To measure the duration and intensity of glottal closure, after listening to each word, the intervocalic part with no waveform was selected and measured (Figure 1).

gottal closure duration (ms)

Fig. 1. GC duration in intervocalic environment [e + u] in [gerje^u]1 'crybaby', FS2, OCW

2.4.2. Creaky voice (CV)

CV was recognized by slight irregularities in glottal pulses, and a considerable decrease in the amplitude (Figure 2) which was probably due to an increase in the glottal stiffness (Ladefoged & Maddieson, 1997: 76-77) [15]. CV duration and intensity were measured as in part 2.4.1 and its F0 was measured by using pitch tract on the selected segment.

pitch line

Fig. 2. CV waveform in intervocalic environment [i + e] in [bi^er=de]j 'indecisive', MS1, OCW

2.4.3. Completely irregular vibration of vocal folds (CIVV)

CIVV showed gradually descending amplitude and occurred when glottal pulses were completely irregular; as a result no F0 could be obtained from the software. The interrupted pitch line in the spectrogram illustrates this fact (Figure 3). Duration and intensity measurement methods were the same as for part 2.4.1.

Fig. 3. CIVV waveform in intervocalic environment [i + i] in [bi^im^n]1 'faithless', MS2, OCW

2.4.4. Vowel-to-vowel transition (VVT)

VVT happened when no consonant existed between two vowels. It reflected simply the waveform changes that could be associated with the changes in vowels' formant frequencies (Ladefoged & Maddieson, 1997: 76) [15] (Figures 4 & 5). In sequence of vowels that vowels' tongue positions were considerably different, e.g. [=] & [i], more time needed to reach the desired position to articulate the second vowel. This period appeared as F2 transition in the spectrogram (Figure 5).

Fig. 4. VVT in [e + i] environment in [x=ne^i]2 'a house', MSi, OCW

F2 transition from [=] to [i]

Fig. 5. F2 transition from [=] to [i] in [^>=^>in]1 'ritual', FS2, ICW

3. Results

Since the main objectives of this study were to investigate the possibility of GS presence in intervocalic positions, in different morphological boundaries and in original simple words, and also determine the acoustic characteristics of its allophones (in case of its presence) in the aforementioned environments, it was supposed that all measurements were allocated to these issues. However, since VVT was observed in a considerable number of data, the amount of this pattern was also calculated. In hiatus, epenthetic consonants, such as [j, w, v, h], V1/V2 deletion, and compensatory lengthening were also observed in a small number of data, which could not be discussed in this paper.

The measurements in Tables 2 to 4 are in accordance with the sequence of acoustic changes in the spectrogram. As Table 2 illustrates, in both groups, glottal closure did not occur immediately between vowels' regular waveforms, but it occurred between a range of completely irregular glottal vibrations (CI) and gradually descending and ascending intensity. These are depicted in Figure 6.

Table 2. F0, Intensity and duration of GC

Speakers Fo (Hz) Intensity (dB) Duration (ms)

Vi CI GC CI V2 V1 CI GC CI V2 CI GC RB CI

Females 197.2 - - - 188.1 69.6 63.3 55.7 64.5 69.6 0.051 0.018 0.001 0.042

Males 116.4 - - - 122.3 67.2 57.7 51.9 57.2 66.6 0.041 0.016 0.001 0.028

0.053 ms

0.071 ms

Fig. 6. Completely irregular waveforms on both sides of GC in intervocalic environment [=+a] in [b=^>adab]1 'polite', FSi, OCW

decreasing F0 of CV compared to that of its surrounding vowels in both groups. Table 5 displays the mean duration of F2 transition in VVT in both groups of speakers.

Table 3. F0, Intensity and duration of CV

Speakers F0 (Hz) Intensity (dB) Duration (ms)

__Vi CV V2 ~ Vi CV V2 CV

Females 198.5 151.5 180.7 66.4 65.5 78.7 0.042

Males 119.4 109.4 113.8 69.8 62.1 68.4 0.032

Table 4. F0, Intensity and duration of CIVV

Speakers F0 (Hz) Intensity (dB) Duration (ms)

__V1 CIVV V2 _ V1 CIVV V2 ~__CIVV

Females 191.1 - 187.4 69.5 63.2 78.9 0.073

Males 113.8 - 116 67.9 58.8 66.8 0.063

Table 5. F2 Transition Duration

Speakers_F2 Transition Duration (ms)

Females 0.088 Males 0.057

In investigating the possibility of the aforementioned phonetic representations occurrence in intervocalic positions, among 684 OCT from all four speakers, GC was observed in 22 (3.21%) tokens, CV in 54 (7.89%) tokens, CIVV in 96 (14.03%) tokens, and VVT in 424 (61.98%) tokens (Table 6). In studying ICT, among the total 276 tokens articulated by the four speakers, no GC was observed in any intervocalic environment; CV was seen in 13 (4.70%) tokens, CIVV in 9 (3.25%) tokens, and VVT in 220 (79.71%) tokens (Table 7).

Table 6. The possibility of GS phonetic representations & VVT occurrence in intervocalic environments in different morpheme boundaries (MB)

and original simple words, OCT

MB & SW GC CV CIVV VVT

FSs MSs FSs MSs FSs MSs FSs MSs

DP + base 21.42% 2.38% 19.04% 35.71% 54.76% 57.14% 4.76% 4.76%

IP + v 4.76% 2.38% 9.52% 8.33% 3.57% 5.95% 75% 72.61%

base + DS 3.70% 1.58% 11.11% 0% 9.25% 22.22% 75.92% 75.92%

base + C 1.33% 0.66% 5.33% 1.33% 8% 7.33% 61.33% 66%

SW 0% 0% 0% 0% 0% 8.33% 100% 91.66%

Total 4.97% 1.46% 8.77% 7.01% 12.57% 15.49% 61.40% 62.57%

Table 7. The possibility of GS phonetic representations & VVT occurrence in intervocalic environments in different morpheme boundaries (MB)

and original simple words, ICT vs. OCT

MB & SW ICT & GC CV CIVV VVT

OCT FSs MSs FSs MSs FSs MSs FSs MSs

IP + v ICT 0% 0% 0% 0% 0% 0% 100% 100%

OCT 16.66% 5.55% 16.66% 33.33% 5.55% 5.55% 61.11% 55.55%

base + DS ICT 0% 0% 20% 0% 0% 0% 80% 93.33%

OCT 0% 0% 10% 0% 0% 10% 90% 90%

base + C ICT 0% 0% 7.69% 1.28% 3.84% 5.12% 71.79% 69.23%

OCT 1.28% 0% 5.12% 0% 0% 0% 66.66% 67.94%

SW ICT 0% 0% 0% 0% 0% 16.66% 100% 83.33%

OCT 0% 0% 0% 0% 0% 8.33% 100% 91.66%

Total ICT 0% 0% 8.69% 0.72% 2.17% 4.34% 79.71% 79.71%

OCT 2.89% 0.72% 7.24% 4.34% 0.72% 3.62% 73.91% 73.18%

4. Discussion

The goals of this study were to investigate the possibility of GS presence in intervocalic environments in both morpheme boundaries and in simple words and to determine its acoustic correlates. In previous studies, GC and CV had been considered as phonetic representations of GS in intervocalic positions. In this study, one other allophone of GS was determined as CIVV (Table 4). Although the majority of previous studies had shown GC and CV presence in intervocalic positions to resolve hiatus (except in fast or colloquial speech), this study illustrated that GS exists mostly in 'DP + base' environment (11.9% GC, 27.38% CV and 55.95% CIVV); in other morpheme boundaries and in simple words, the considerable number of data represented VVT in all speakers, 77.30% in OCT and 79.71% in ICT (Tables 6 & 7).

Phonetically speaking, since the majority of results illustrated no consonant presence/insertion in intervocalic positions except in 'DP + base' environment, only a part of the hypothesis based on the presence/insertion of GS between two adjacent vowels in morpheme boundaries and in original simple words can be confirmed. According to these findings, it can be concluded that:

1. In standard formal Persian, the absence of GS (with its three phonetic representations) in intervocalic environments, in morpheme boundaries (except in 'DP + base'), and in original simple words is much more probable than its presence.

2. In addition to the previous findings on the possibility of GS deletion in intervocalic positions in fast and colloquial speech, it is also possible to omit all three phonetic representations of GS in intervocalic environments and have VVT in standard formal Persian.

Acknowledgements

The author would like to express her deep sense of gratitude to her supervisor, Dr. Golnaz Modarresi Ghavami, Assistant Professor at Allameh Tabataba'i University, Department of Linguistics, for her invaluable help and support in all stages of the research.

References

Sadeghi, A. A. (2001). Hiatus and mediator consonants issue in Persian. In Persian historical issues (pp. 25-49). Tehran: Sokhan Publishing.

Boersma, P & Weenink, D. [Praat software]. http://www.fon.hum.uva.nl/praat/

Casali, R. F. (1996). Resolving hiatus. (Doctoral dissertation). University of California, USA.

Lazard, J. (2005). A grammar of contemporary Persian (M. Bahreini, Translation & H. Milanian, Descriptions). Tehran: Hermes Publishing. (Original work published 1957)

Nye, G. E. (1954). The phonemes and morphemes of Modern Persian: A descriptive study. (Doctoral dissertation). University of Michigan, USA. Stagi, A., Namvarfargi, M. & Keramati Yazdi, S. (2011). An acoustic analysis of glottal stop and the study of the possibility of existing two

adjacent vowels in successive syllables in rapid speech in Persian. Comparative language and literature research, 1(4): 27-50. Yarmohammadi, L. (1964). A contrastive study of Modern English and Modern Persian. (Doctoral dissertation). Indiana University, USA. Haghshenas, A. M. (2003). Hamze in Persian language and orthography. In Persian language and literature at the crossing between tradition and

modernity, 267-280. Tehran: Agah Publishing. Lambton, A. K. S. (1971). Persian grammar. London: Cambridge University Press.

Haghshenas, A. M. (1978). Dual roles of hamze in Persian phonetic structure. Journal of the academy of Persian language, 3: 23-57.

Windfuhr, G. L. (1979). Persian grammar: History and state of its study. The Hague, Paris & New York: Mouton Publishers.

Jahani, C. (2005). The glottal plosive: A phoneme in spoken Modern Persian or not?. In E.A. Csato, B. Isaksson & C. Jahani (Eds.), linguistic

convergence and areal diffusion: Case studies from Iranian, Semitic and Turkic (pp.79-96). London & New York: RoutledgeCurzon. Parmun, Y. (2002). A phonological study of contemporary standard colloquial Persian: A generative, metrical, and lexicalist approach.

(Doctoral Dissertation). Tarbiat Modarres University, Iran. Kalbassi, I. (1992 [2008]). The derivational structure of word in Modern Persian (3rd ed.). Tehran: Institute for Humanities and Cultural Studies. Ladefoged, P. & Maddieson, I. (1997). The sounds of the world's languages. Blackwell Publishing. Modarresi Ghavami, G. (2012). Phonetics: The scientific study of speech. Tehran: SAMT.