Welcome to Scribd. Sign in or start your free trial to enjoy unlimited e-books, audiobooks & documents.Find out more
Download
Standard view
Full view
of .
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
Punjabi Phonemic Inventory

Punjabi Phonemic Inventory

Ratings: (0)|Views: 175|Likes:
Published by crazylinguists

More info:

Published by: crazylinguists on Aug 16, 2012
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

08/16/2012

pdf

text

original

 
 179
PHONEMIC INVENTORY OF PUNJABI
 
NAYYARA KARAMAT 
ABSTRACT
Punjabi belongs to Indo-Aryan family. In thispaper its phonemic inventory is discussed.Dialect spoken in Lahore region isconsidered. Documented inventories wereconsidered as basis of the analysis and thenfurther experiments are conducted for investigation of status of the phonemesparticular to the dialect.
1. INTRODUCTION
Punjabi is an intercontinental language. Itsspeakers are spread all over the world. Thenatives of the Punjab region of India andPakistan speak this language.Punjabi is a member of the Indo-Aryanlanguage family (See Appendix A) [6].Punjabi in India is more open to Sanskritand traditional sources whereas Punjabi inPakistan is influenced by Perso-ArabicSources [7].Punjabi language has a number of regionaldialects that are spoken in both India as wellas in Pakistan. The main dialects of Punjabiin India are Majhi, Doabi, Malwai andPuwadhi. The regional dialects likePothohari, Lehandi & Multani are spoken inPakistan [7]. The dialects of different areashave been observed to have distinctdifferences in their phonemic inventories.In India the Punjabi is written in Gurmukhiscript as far as in Pakistan it is written inPerso-Arabic Script. Gurmukhi is consideredmore suitable script for writing Punjabi as itwas devised to write Punjabi by a Sikh Guru Angad [13].
2.
LITERATURE REVIEW &PROBLEM STATEMENT
 
The phonemic inventory of Punjabi isdiscussed in many books, but most of thework is done on the dialects spoken inEastern Punjab. Due to difference of religious backgrounds, Punjabi spoken inEastern Punjab is more influenced bySanskrit where as Punjabi spoken inWestern Punjab is influenced by Persianand Arabic. This difference causes somechanges in the phonemic inventory of Western Punjab. Dialects spoken in WesternPunjab also differ in their phonemic system.This paper presents the phonemic inventoryof the Punjabi dialect spoken in Lahore andits surroundings. After collecting phonemic inventory fromdifferent sources, my aim was to verify themfor my dialect.One of the problems was to check for theexistence of [
 
], a palatal nasal. There werereferences to the existence of a palatalnasal /
 
/ in various inventories of Punjabi(Colin, 1991), [7], but these did not mentionthe dialect or considered Majhi dialact(spoken in Amritsar) as the standard.Gurmukhi orthography [8] also shows thatthere is a distinguished palatal nasal inPunjabi, particularly in context of palatalaffricate /
d
/ and diphthong /
a
/, nasalsound is written differently as compared toother contexts [9], [10].For example, in
/
d
nd
/
 
and
/
nani
     
/,
 
is used for nasal occurring in /
d
/
or 
 /
a
/
   
context. Whereas for alveolar nasalis used [9], [10]. All the sources claimed three short vowels inPunjabi (Colin, 1991), [7]. But it wasobserved that quality of /
/ and /
/ shortvowels changes with stress. It seems thatthey are converted to short form of vowels/
e
/ and /
o
/ respectively. This phenomenonmay increase number of short vowels to five.The second problem was to examine theexistence and phonemic status of these twonew vowels.
 
Center for Research in Urdu Language Processing180
Other controversial consonants were /
 
/and /
/. Their phonetic characteristics had tobe examined to check whether these areretroflex lateral and retroflex nasal or flaps.Punjabi is a tonal language (Colin, 1991),[7]. Tonal languages employ patterns of pitch variation to distinguish betweendifferent meanings of word that have thesame pattern of consonants and vowels.Tone features in tone languages aresegmental and phonemic in function(Pickett, 1999). One of the problems was toobserve the tonal patterns occurring inPunjabi, which cause lexical differences inwords.
3. METHODOLOGY
For confirmation of phonemic existence of sounds I used parallel distributions of thosesounds. I tried to find minimal pairs for allthe sounds that might be allophonic to eachother.See Appendix B for a list of all theinvestigated pairs and the ones that werefound to be minimal.To investigate the phonetic characteristics of sounds /
 
/ and /
/, I took minimal pairs of /
l
/and /
n
/ with their retroflex variations. Pairsused for this study were as follows:
[
m
n
] [
m
 
][
kana
] [
kaa
][
v
l
] [
v
][
k
l
] [
k
]
To verify existence of palatal nasal, I choosefour words in _ 
d
and _ 
a
 
context. Thewords were chosen from a GurmukhiDictionary [9], [10] since Gurmukhi script isphonemically more directly linked to spokenPunjabi than Urdu script. The words used for this analysis are as follows:/
d
nd
/
marriage party/
m
ndi
/
bed
 /
nani
/
girl/
d
na
/ parsley like garnishTheir acoustic properties were analyzed todistinguish alveolar nasal from palatal nasal.Vowel transitions while entering into thenasal and at the release were observed.Since the distinguishing acoustic featureswere not found sufficiently documented [11],[12], an additional experiment wasconducted to verify them.It was documented that /
 
/ can bedistinguished by the feature [+compact] [11]i.e. the power of the signal is located in themiddle, which implies that F2 is closer to F1[12]. To verify this, a vowel with high F2 (
)was selected, and another with low F2 (
u
)was selected. The contexts /C
/ and /C
u
/were then recorded (C denoting aConsonant), where C was [
n
] and [
 
], andthe rate of transition of the F2s wasobserved. It was expected that in /C
u
/context, rate of transition of F2 will begreater for /
n
/ than /
 
/. In /C
/ contexthigher rate of transition for /
 
/ is expected.To check the existence of shorter form of vowels /
o
/ and /
e
/ some pairs of wordscontaining /
/ and /
/ were taken. One of which is suspected to have change in vowelquality resulting in change to /
o
/ or /
e
/.Words used for this experiment are asfollows:/
t
/
/
tan
/
 
/
pr
/ /
prki
]/
til
/ /
tla
]/
sr
/ /
srili
//
mu
/
/
mui
/
 Words listed in first column are expected tohave change in vowel.To observe the tonal structure of Punjabi,some words containing different tones wererecorded. Punjabi orthography in Urdu or Gurmukhi script does not show tones. It wasobserved that orthographically voicedaspirated stops are converted intocorresponding voiceless stop with high toneon the syllable (Khan, 1997). Moreover,tones are observed on the words containing
 
 181
/h/ in middle and end of the word in Urduorthography. The word selection was madein accordance with these observations. For words selected for tonal analysis see Appendix C.For all experiments, words were recordedfive times in random sequence by five nativespeakers of Punjabi (Lahore accent).
4. RESULTS
While comparing spectrograms of 
/
 
/
and/n/, a significant difference can be seen. In
/
 
/
there is very small duration of the closureperiod and falling F3, which shows it is aretroflex flap (Pickett, 1999). Another argument for considering it a flap is speaker cannot hold it while articulating
/
 
/
. In thesame way the spectrogram of /
/ shows aslight retro flexion by falling F3. See Appendix D for spectrograms.For the /
 
/ experiment it was verified that thedocumented features were correct.F2increase in /C
/ context is more for /
 
/ andits decrease in case of /C
u
/ context is morefor /
n
/. See Appendix E for spectrograms.While observing formants of nasal withpalatal consonant, it is observed that F2 isrising while entering in the nasal and comingdownward at release. This shows that nasalin the palatal context does not getpalatalized. See Appendices F.For the short vowel experiment, theformants of the documented /
/
and
/
/
in suspected words were found to bedifferent as expected; their F2s were foundto be decreasing and F1s were increasing.Their length of duration was also increasing.For the spectrograms see Appendices G.Even though the length of duration wasincreasing, they were not being changedinto long vowels, since the long vowels wereof even longer duration.The minimal pairs
[
ser
]
unit of measurement[
ser
]
(short e) headprove that the long and short forms of 
[
e
]
 are phonemic, but the short form of [
e
] isallophonic with [
]. And the minimal pairs[
por
] morning / crumbs[
por
]
(with short [o]) to crumble
 
prove the same for the other set.From the experiments on the tonal structureof Punjabi three tones high, mid and lowwere found. While producing these tones,there is no blockage of air in the mouth.They are pronounced concurrently with asyllable.High tone is higher then the other two tonesand the syllable with this tone is also shorter than those with other two tones.Mid tone is considered to be an intermediatein pitch between the high and the low tones.The syllable is of an intermediate height.Low tone has been described as the lowestof the three tones. The pitch tends to fall.The syllable under this tone is longer incomparison with the other two. These threetones mark lexical contrast. Some examplesare as follows:
/
koa
/
horse
/koa
/
whip
 /
koa
/
leper 
/
ta
/
peep
/
ta
/
enthusiasm
/
ta
/
teaSee Appendix C for an example of tonalminimal pair.The conclusion drawn from the results of theexperiments stated above are as follows.Punjabi system of sounds involves fivedistinctive tongue positions: labial, dental,retroflex, palatal and velar. The retroflexposition varies from person to person andmay involve curling back of the tongue tomake the contact with the underside of thetip, or merely retraction.

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->