Arabic Dialects Thread

A forum for guides, lessons and sharing of useful information.
Post Reply
Davush
sinic
sinic
Posts: 412
Joined: Sat 10 Jan 2015, 14:10

Arabic Dialects Thread

Post by Davush » Sun 19 Nov 2017, 16:06

As Shemtov mentioned, I thought it would be a good idea to start a thread on the Arabic dialects. I will try and give an overview of the main features of each group. I am most familiar with Norther Gulf dialects, so if anybody is more knowledgeable on Levantine/Egyptian/North African please do contribute!

Gulf Arabic
This is the dialect I am most familiar with (specifically Kuwaiti), and most of the features are shared among all the gulf accents (Kuwait, Bahrain, Qatar, UAE, parts of Eastern Saudi and Oman to a lesser extent). Many of the features are also found in Southern Iraqi.

Vowels

/i u/ collapse into /ǝ/ in the Gulf and also in northern Saudi, although both /ɪ ʊ/ occur as realisations of /ǝ/ they are conditioned by surrounding consonants. /ʊ/ is only very slightly rounded, and lack of a true short /u/ is quite characteristic.

e.g. /kutub/ > /kǝtǝb/

/a/ > Gulf /ɑ æ/ - This actually happens in MSA as well. /ɑ/ is an allophone near emphatics. Unlike other dialects, however, it is more often /æ/ near /q ħ ʕ/. It often becomes /ǝ/ when unstressed. /æ/ can be closer to /ɛ/ sometimes.

/ʕala/ > /ʕælæ/
/ðˤall/ > /ðˤɑɫɫ/

Word finally /æ/ is often /ɛ/ or even /e/ (especially in the northern half of the gulf):
/ahlan/ > /hala/ > /hæle/

MSA /a:/ is backed to /ɑ:/ in all positions in the Gulf, this is very characteristic of the Gulf in that is basically doesn't have imāla (raising of a/a:). In Bahrain it is often even rounded to /ɔ:/ and sometimes in parts of UAE as well.

MSA: /sala:m/ > /sælɑ:m/
MSA: /ba:b/ > /bɔ:b/

/aj aw/ become /e: o:/ which is common throughout the Arab dialects. In the Gulf they may be slightly lower /ɛ: ɔ:/

/bajt/ > /bɛ:t/
/fauq/ > /fɔ:g/

Gulf Arabic doesn't shorten vowels before consonant clusters unlike most of the Levant and Egypt.

Consonants

I will list the main changes and characteristics of Gulf Arabic.

/θ ð/ are retained in the Gulf and show no change.

/q/ > /g/ is probably the most obvious. It happens more consistently than other dialects (for example some Levantine dialects have this change, but it doesn't seem to be applied nearly as much as in the Gulf, and it also has a certain stigma in the Levant which it doesn't in the Gulf.) Higher register words are less likely to have /g/.

/qahwa/ > /gahwa/
/qa:la/ > /ga:l/

/g/ > /ʤ/ is very peculiar to the Gulf. It usually happens near front vowels.

/qari:b/ > /gari:b/ > /ʤri:b/
/muqa:bil/ > /mga:bil/ > /mʤɑ:bil/

Saudi dialects are less likely to have this change, keeping /g/.

/k/ > /ʧ/ is also very common. It supposedly happens near front vowels, but it actually seems quite random. Some parts of the Levant also have this, but it seems to have a bit of a stigma there.

/kabi:r/ > /ʧbi:r/
/kam/ > /ʧam/

but

/kaθi:r/ > /kθi:r/

The feminine 2p suffix -ik also becomes /ǝʧ/:
/kutubuki/ > /kǝtǝbǝʧ/

/ʤ/ > /j/, probably motivated by /g/ moving to /ʤ/. This is also quite confusing for people not from the Gulf. It hasn't affected as many words as the previous two changes, however. In UAE and parts of Oman and Yemen it can be closer to /ɟ/.

/ʤumʕa/ > /jǝmʕa/
/ʤanb/ > /jamb/ > /jamm/

Pharyngealisation
/l m b r f/ > pharyngealised equivalents /ɫ mˤ bˤ rˤ fˤ/. The Gulf has a strong pharyngealisation of these, sometimes even when other pharyngeals aren't around. Pharyngealisation manifests itself on the consonant in the Gulf. In other dialects, vowel change is often the main indication for it, but possibly because of the reduced vowel system and /ɑ:/ for /a:/ in all positions, the pharyngealisation is strongly heard on the consonant. Pharyngealized /m b/ can also have simultaneous labialization. Also /ʕ/ seems to be closer to a stop than an approximant/fricative.

/dˤ/ and /ðˤ/ are merged into /ðˤ/ (unlike elsewhere where it is often /zˤ/.)

/qalb/ > /gaɫbˤ/
/umm/ > /ǝmˤmˤ/
/ummi/ > /ǝmˤʷmˤʷi/
/ðˤill/ > /ðˤǝɫɫ/
/ʕira:q/ > /ʕǝrˤɑ:g/


Syllable Structure
Gulf Arabic tends to avoid final consonant clusters, splitting them up.

/ʔakl/ > /ʔakǝl/
/baħr/ > /baħar/

Some common clusters like /nt/ are common, though. Maybe it has something to do with the sonority hierarchy. /bǝnt/ remains /bǝnt/. Iraqi tends to avoid all final clusters, so /bint/ > /binit/, etc.

The pattern CVCVCV often becomes CCVCV which is quite different from other dialects, especially the feminine past form of verbs. Guttural consonants have a tendency to move before the vowels.

/katabat/ > /ktǝbat/ (usually /katbat/ in other dialects).
/qahwa/ > /gahwa/ > /gᵊɦæwe/
/aħlǝf/ > /aħálǝf/
/aʕrˤǝf/ > /aʕǝrˤfˤ/

If an initial cluster is created, epenthetic /ǝ/ can be inserted.

Unlike Egyptian, but similar to Levantine, words like /madrasa/ are stressed on the first syllable.


Grammar/Morphology
Gulf Arabic is quite conservative in some ways.

The final -n of the plural present forms is kept. This is lost in most other dialects.
/tǝktǝbi:n/ 'you (f.) write'
/jǝktǝbu:n/ 'they write'
etc.

The Southern Gulf and parts of Saudi have kept the fem. plural forms in -in or -an for 3p and -tin for 2p.

/ktǝbǝn/ 'they (f.) wrote
/kǝtabtǝn/ 'you (f.) wrote'

Some parts of Saudi have the vestige of a morphological passive. This doesn't really occur in the Gulf states though, which prefer the /in-/ form.

/kítab/ 'he wrote'
/ktib/ 'it was written' ( /inkítab/)

/ðibaħ/ 'he slaughtered'
/ðbiħ/ 'it was slaughtered (/inðíbaħ/)


Some speakers also still have indefinite nouns in /-n/. Tanwīn has been lost everywhere else. I mostly only hear this when it's followed by an adjective.
/bju:tǝn ʧbi:ra/ 'big houses'

The 3p. object/possessive suffix is always /ah/ in the Gulf. Nearly everywhere else has /uh/ or /oh/ instead. Possibly motivated by the general avoiding of short /ʊ/?

/bɛ:tæh/ 'his house'
/ʃǝftæh/ 'I saw him'


Vocabulary
Gulf Arabic has quite a lot of vocabulary peculiar to it. Most of this is Southern Iraqi too. Here are some words I can think of. Often the word for 'what' is a big indicator as to dialect.

/ʃǝnǝw/ what? (often contracted to simply /ʃ/)
/mǝn/ who?
/χɔ:ʃ/ good
/jɑ:hǝl/ child (from jāhil 'an ignorant one')
/ʕɔ:d/ big, old
/rajjɑ:l/ man
/jabi/ to want
/haga/ to think, believe
/tǝħaʧʧa/ to speak
/ʕajal/ well, right, so
/zɛ:n/ good
/dǝʃʃ/ to enter

/zɛ:n ʕajal, mǝn jabi jǝtħaʧʧa χali:ʤi wijjɑ:j?/ [:D]

Overall, there is an idea that Gulf Arabic is quite conservative and close to MSA, but not-watered-down Gulf Arabic can be quite difficult for other Arabs to understand because of the sound changes in addition to archaic vocabulary & Persian loan-words. It is associated with 'Bedouin culture' in Egypt etc to a degree.

To be continued.
Last edited by Davush on Sun 19 Nov 2017, 17:52, edited 7 times in total.
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4472
Joined: Sun 31 Mar 2013, 12:16

Re: Arabic Dialects Thread

Post by DesEsseintes » Sun 19 Nov 2017, 17:06

This is amazing! I love Arabic, and I really look forward to seeing more. [:D]
User avatar
Shemtov
mayan
mayan
Posts: 2159
Joined: Mon 29 Apr 2013, 03:06

Re: Arabic Dialects Thread

Post by Shemtov » Sun 19 Nov 2017, 18:02

Does Gulf Arabic include Yemenite? Because you don't list Yemen as a country that speaks Gulf Arabic, Wiki lists as a separate dialect, and yet you bring an example of "Yemenite Arabic". I would imagine it to be quite different, because I know in the Pre-Islamic period Yemen spoke South Semitic languages, as opposed to the Central Branch that Arabic is in, I would think that they would have had an influence on the Arabic spoken there.
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien
Davush
sinic
sinic
Posts: 412
Joined: Sat 10 Jan 2015, 14:10

Re: Arabic Dialects Thread

Post by Davush » Sun 19 Nov 2017, 18:04

Shemtov wrote:
Sun 19 Nov 2017, 18:02
Does Gulf Arabic include Yemenite? Because you don't list Yemen as a country that speaks Gulf Arabic, Wiki lists as a separate dialect, and yet you bring an example of "Yemenite Arabic". I would imagine it to be quite different, because I know in the Pre-Islamic period Yemen spoke South Semitic languages, as opposed to the Central Branch that Arabic is in, I would think that they would have had an influence on the Arabic spoken there.
Yes sorry Yemeni is quite different. I only mentioned /ʤ/ > /ɟ/ there because certain Yemeni dialects share that with UAE/Oman. The main odd features that I think of in some Yemeni dialects are /am/ instead of /al/ as the article, and /k/ instead of /t/ as the 1p marker. E.g. /katabk/ instead of /katabt/. I'll try and write more about Yemeni in another post.
User avatar
Shemtov
mayan
mayan
Posts: 2159
Joined: Mon 29 Apr 2013, 03:06

Re: Arabic Dialects Thread

Post by Shemtov » Wed 22 Nov 2017, 01:38

Davush wrote:
Sun 19 Nov 2017, 16:06



/haga/ to think, believe


/haga/ also meant"thought" in Palestinian Jewish Aramaic, but a. I would expect that to show up more in Levantine, and especially in Palestinian, Arabic, not Gulf Arabic and b. giving the sound correspondences between Aramaic and Arabic, I would have expected such a loan to be /had͡ʒa/ around Qurʾānic times and thus /haja/ in Gulf Arabic.
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien
Davush
sinic
sinic
Posts: 412
Joined: Sat 10 Jan 2015, 14:10

Re: Arabic Dialects Thread

Post by Davush » Wed 22 Nov 2017, 10:12

Very interesting! Although I have only seen it spelled as هقى indicating it comes from /haqaa/. The present form is yihaga and the verbal-noun is hagwa (maybe indicating earlier haqawa?) Perhaps it’s just a coincidence? Or perhaps a borrowing which retained /g/ although that does seem a bit unlikely.
Davush
sinic
sinic
Posts: 412
Joined: Sat 10 Jan 2015, 14:10

Re: Arabic Dialects Thread

Post by Davush » Wed 22 Nov 2017, 16:08

Yemeni Dialects

This is really a dialect group in itself although there is a fairly obvious split between north and south Yemeni. By north I am usually referring to San’ani and South the variety spoken around Ibb and Ta’izz. Aden and the south coast also have some peculiarities. I really like Yemeni Arabic but I find it hard to pinpoint what makes it so unique. For me, the stress patterns and intonation make it stand out as it doesn’t have as dramatic sound changes as say the Gulf or Egypt does. Also the Yemenis I know probably speak some levelled-out form of the dialect or koine because they come from various places.

Yemeni dialects generally preserve the three short vowels /a i u/ better than the Gulf. Some western Saudi dialects also do. The distribution doesn’t seem 100% the same as MSA though, as /i u/ are sometimes conditioned by surrounded consonants. When unstressed they may reduce to schwa. /i u/ are closer to /ɪ ʊ/ as is usual for Arabic short vowels.

Example would be /iskut/ meaning ‘be quiet’ which is /iskit/ or /ǝskǝt/ in the Gulf.

/a/ in San’ani seems to be closer to a true /a/ than elsewhere although /æ~ɛ/ can be heard.
Word final /ah/ seems to be raised to /ih/ in Yemen, with the /h/ often being audible, whereas in almost evervywhere else it is now simply /a/ (and its allophones).
/xʊbzɪh/ a loaf of bread

The dipthongs /aj aw/ seem to usually remain as diphthongs in Yemeni dialects.

Long /a:/ shows raising, unlike the Gulf, but not to the extent of the Levant. When not near pharyneals it is probably closer to /ɛ:/. When near pharyngeal it is backd to /ɑ:/. The Gulf is actually probably the odd one out having /ɑ:/ in all positions. Maybe this is to do with Persian influence?

Most Yemeni dialects have /q/ > /g/. This seems to be very strong in Yemen as for example when some of my friends read texts written in MSA, they will still have /g/ where most other speakers would have /q/ when reading. It has been reported that some traditions also have /g/ for /q/ in Qur’anic recitation. The Ta’izz area is known for /q/ remaining /q/.

I should have mentioned that some Gulf dialects have /q/ > /ʁ/ and /ʁ/ > /q/ effectively reversing ghayin and qaaf. Some Yemeni dialects have this too. I can’t remember where I read it, but apparently the ghayin/qaaf confusion is an old phenomenon and certain tribes have evidently retained that.

/ʤ/ seems to show some variation. The most common realization I have heard is /ɟ/ which can sometimes be closer to /ʤ/ - the further south in Yemen, the closer to unpalatalized /g/ is seems to become.
Wikipedia writes that the Tihami dialect has /q/ as /q/ and /ʤ/ as /g/ ‘like most other Yemeni dialects’, but I think this is inaccurate. It seems that these are features of dialects on the southern coast, including Aden. But inland has the more usual /q/ > /g/ and /ʤ/ > /ʤ~ɟ/

Northern Yemeni has /dˁ ðˁ/ > /ðˁ/ whereas Southern has them as /dˁ/.

Emphatic /l/ isn’t common as it is in the Gulf. I.e. /qalb/ > either /qalb/ or /galb/ but not /galˁbˁ/.

Stress is generally very strange in Yemeni Arabic, at least to those used to other dialects, and I think it is one of the things which throws people off. Often the first syllable is stressed regardless, so things like:
máwjuud vs mawjúud
ħáaðˁiriin vs ħaaðˁiríin

San’ani Arabic sometimes voices /t tˁ/ intervocalically which is quite unique.
/laa: tinsaaʃ/ > /laa dinsaaʃ/ ‘don’t forget’ (Gulf: /laa tǝnsa/)

There also seems to be some diphtongization of /i: u:/ in final syllables, but I'm not exactly sure when/how common this is. I have heard people say /balqajs/ or /balqejs/ for the name /balqiis/ for example.

Morphology

Morphological features are probably the more prominent features of Yemeni Arabic as they can be quite different from everywhere else.

In the north, it is common for the 1p past ending to remain –tu whereas everywhere else it is –t:
South-west has –k instead (but I think this is probably being replaced by -t among younger speakers - I know several people from Ta'izz and I have heard this probably only once or twice, that said, they may have been speaking with people from outside the region).

qultu > gultu (Gulf: gǝlt)
qultu > qulk katabtu > katabk

Some areas of the south-west also have /ʃ/ for the 2p. fem. Past whereas elsewhere as –ti or –tay:
Katabti > katabʃ

I think the 3p. f. past form can also be different from elsehwere in that some Yemeni dialects have -/ah/ or -/an/ instead of /at/:
katabat > katabah / kataban (Gulf: ktibat)

In San'ani, I think the 1p. object suffix is -na, and the 1pl. suffix is -iħna (compared to elsewhere: -ni and -na).

This may have come from –ki > ʃ. Even San’a as -ʃ as a 2p. fem. Possessive suffix (but not in the verb form):
kitaabuki > kitaabiʃ

The feminine plural forms are well retained in Yemeni.
anti (2p. f. sg.) antan (2p. f. pl.)
hii (3p. f. sg.) han (3p. f. pl.)

It might be easier to visualize if I just present a fully conjugated verb in Classical, Gulf, and Coastal Yemeni (although please bear in mind this is what I remember from books as the Yemeni I hear is mostly not of this kind):

Qaala ‘to say’
MSA > Gulf > Southern Yemeni / San’ani Yemeni

Qultu > gilt > qulk / gultu
Qulta > gilt > qulk / gult
Qulti > giltay > qulʃ (?) / gulti
Qaala > gaal > qaal / gaal
Qaalat > gaalat > qaalat / gaalat
Qulnaa > gilna > qulna / gulnaa (retains long vowel)
Qultum > giltaw > qulkum / gultu
Qultunna > giltaw (southern gulf may have giltin) > qulkun (?) / gultinna
Qaaluu > gaalaw > qaalu / gaalu
Qulna > gaalaw (southern gulf may have gilin) > qaaleen / gulna

As far as I can tell from what I’ve read and heard, the northern dialects (including San’a) while still distinct, tend to have more common features shared with the south of the Arabian peninsula, while the coastal dialects have the truly strange stuff including /am/ as the definite article.

Future Markers
An interesting feature that a lot of Yemeni dialects share is the use of /ʃa/ for the future, which has been said to be derive from the classical Arabic particle /sawfa/ or prefix /sa/. Most other dialects have /ħa/ from /raajiħ/ (going to). I forgot to mention in the previous post that the gulf uses the prefix /b/ to indicate future. This can confuse other speakers where /b/ is used to indicate a habitual/ongoing action.

The Gulf future /b/ and Levantine/Egyptian habitual /b/ actually have different origins:
Gulf /b/ comes from the verb /baʁaa/ ‘want’, and Egyptian/Levantine /b/ presumably from the preposition /bi/ ‘in/at/on’.

The negative is formed by /maa/ + /ʃ/. This is quite common throughout the dialects. The Gulf however does not have the negative /ʃ/ suffix
/maa katabtiʃ/ ‘I did not write’

Vocabulary
Vocabulary is also where Yemeni differs quite sharply. Notably many dialects have retained /maa/ to mean ‘what’. Sometimes with an object suffix: /maahu/ or /maahi/ depending on gender.
/maahu haaða/ ‘what’s this?’ Gulf: /ʃǝnǝw haaðǝ/
‘Where’ is commonly /ajn/ (Classical Arabic /ajna/) Nearly everywhere else has /we:n/ or /fe:n/).

I remember one time I was overhearing some Yemenis speaking and I had to ask if they were actually speaking Arabic, it sounded like Arabic, but the words weren't familiar. It turns out they were, but most the words they had used in that sentence were archaic/rarely used/incomprehensible in comparison with other dialects.
Post Reply