Pronunciation: ـي الـ - قميصي الجديد
- Thread starter Ali.Hm
- Start date
I would never pronounce it long.
No! 
That sounds horrible.
That sounds horrible.
And that would be a wrong pronunciation.
*The final ي of قميصي is pronounced short, not long.
*The ا of الجديد is not pronounced.
Any pronunciation that violates either or both of these two conditions does not sound native.
qamiiSi 'l-jadiid
qamiiSii 'l-jadiid
qamiiSi al-jadiid
qmiiSii al-jadiid
(Of course, if the words are pronounced separately, or with a pause in between, the "a" is pronounced, but the final "i" is still short.)
I don't think this is right. Surely the pronunciation of قميصِ and قميصي are not the same?
They are! I think the only exception is Qur'anic recitation, which follows its own rules. In all other MSA contexts, native speakers pronounce قميصي with a short i at the end. I believe this applies to almost all suffixes ending in vowels; the only exception I can think of is the dual -aa, which is pronounced long.
I agree with elroy - except in the most formal kind of pronunciation, no final vowels really occur long except aa (and that only in the dual, probably lengthened as an affectation to distinguish it, and in nouns like mubaaraah, which often get pronounced with a -t to avoid the awkward long a anyway). This might be different in, say, Moroccan pronunciations of MSA where the stress is final. That transliteration is not good evidence; obviously the final vowels are written long in transliteration because that's how they're written in Arabic. And even if that weren't the case, the transliteration here is p. dubious - bilady is a very different convention from the one that appears in the next line.
From what I have heard and read, most speakers of MSA merge long and short opens vowels at the end of a word, regardless of the following word. In classical grammar, word-final long open vowels are closed, and therefore shortened, by a word-initial consonant cluster in the following word. Thus according to both modern practice and classical grammar, this phrase is pronounced qamiiSi ljadiid.
I am a native speaker and pronouncing it long sounds utterly unnatural to me. I've never heard a native speaker pronounce it that way. Transliterations are hardly reliable evidence; there are numerous transliteration styles and conventions, and they differ with regard to the extent to which they reflect pronunciation vs. orthography. I'm not familiar with Cowan, but if you want to trust one book over the intuition and experience of a native speaker, go right ahead.
Yes, sorry the translation is a bit inconsistent, but you can listen to the song and it is clearly pronounced long lakii hubbiiiiiii wa fuaaaadii!
No, Andyroo. It's pronounced short in لك (which ends in a kasra, not a ي, so why would it be long!) and فؤادي. It's extended in حبي to match the melody of the song, not because it would be pronounced long normally.
I was just using the book to back up what I've always thought to be the case.
Just to be sure we are not getting our wires crossed, would you say that the pronunciation of the two "i"s in the word "بنتي" are the same: binti not bintii?
Yes, they are identical. I don't think any wired are being crossed.
Yes, they are identical. I don't think any wired are being crossed.
Wow - that is really surprising. I wonder if there are dialectical differences (I mainly hear Iraqi). I would have been willing to bet quite a large amount of money that the two "i"'s in بنتي were pronounced differently!
EDIT: But even listening to the song Bilady - in the word bilady itself you can clearly hear the i is pronounced differently to the "y" at the end.
And further, I checked another resource (Ryding). She transcribes يا أرض بلادي as yaa 'arD-a bilaad-ii.
These are reliable resources.
Last edited:
I don't think songs are a good indicator of conventional pronunciation. It's normally "bilaadi."
As far as the transliterations, maybe they are transliterated that way because in certain contexts (like Qur'anic recitation) they can be pronounced long, and maybe they want to make it clear they're not orthographically short vowels (7arakaat).
But here's the thing: If بلادي were to pronounced with a long ii at the end, one of two things would need to happen:
1. The stress would shift to the final syllable, as tends to happen because suffixes with long vowels attract stress. We know, however, that it's stressed on the penultimate syllable, though, so that possibility is ruled out.
2. The stress would remain on the penultimate syllable but the final vowel would still be long. This would be an awkward thing to do and just not natural in Arabic. If I try to do it my pronunciation comes out strange and forced. At least the vast majority of the time - if not all the time - the vowels in unstressed syllables are short, not long.
One could argue that the stress remains on the second syllable in this case because the vowel in that syllable is long. But that argument is easily countered because the stress is not on the last syllable in بلدي (baladi) either, despite the fact that the second vowel is short. This one is really the final nail in the coffin of the theory that the ي is pronounced long. If it were, the stress would absolutely have to shift to the final syllable, because in Arabic if a word has only one long vowel the stress is invariably on the syllable with that vowel. But the stress is definitely not on the last syllable here.
The upshot is that, for whatever reason, in Arabic the vowels in suffixes with long vowels, with few exceptions, are pronounced short. Incidentally, there's a reason ى, which always appears word-finally, is called ألف مقصورة.
As far as the transliterations, maybe they are transliterated that way because in certain contexts (like Qur'anic recitation) they can be pronounced long, and maybe they want to make it clear they're not orthographically short vowels (7arakaat).
But here's the thing: If بلادي were to pronounced with a long ii at the end, one of two things would need to happen:
1. The stress would shift to the final syllable, as tends to happen because suffixes with long vowels attract stress. We know, however, that it's stressed on the penultimate syllable, though, so that possibility is ruled out.
2. The stress would remain on the penultimate syllable but the final vowel would still be long. This would be an awkward thing to do and just not natural in Arabic. If I try to do it my pronunciation comes out strange and forced. At least the vast majority of the time - if not all the time - the vowels in unstressed syllables are short, not long.
One could argue that the stress remains on the second syllable in this case because the vowel in that syllable is long. But that argument is easily countered because the stress is not on the last syllable in بلدي (baladi) either, despite the fact that the second vowel is short. This one is really the final nail in the coffin of the theory that the ي is pronounced long. If it were, the stress would absolutely have to shift to the final syllable, because in Arabic if a word has only one long vowel the stress is invariably on the syllable with that vowel. But the stress is definitely not on the last syllable here.
The upshot is that, for whatever reason, in Arabic the vowels in suffixes with long vowels, with few exceptions, are pronounced short. Incidentally, there's a reason ى, which always appears word-finally, is called ألف مقصورة.
Last edited:
I think your logic falls down when you say the stress has to fall on the long vowel. I don't think that is a rule in Arabic.
For me, there is a clear stress on the first syllable in بنتي even though (I think) the ي is long.
The quality of the two vowels in بنتي are different. It is like the word "mini" in English.
For me, this difference in quality is very apparent in the song - fuaadii bilaadii hubbii the end vowel is very different from the first vowel in bilaadii.
And ألف مقصورة is pronounced long according to Wehr!
For me, there is a clear stress on the first syllable in بنتي even though (I think) the ي is long.
The quality of the two vowels in بنتي are different. It is like the word "mini" in English.
For me, this difference in quality is very apparent in the song - fuaadii bilaadii hubbii the end vowel is very different from the first vowel in bilaadii.
And ألف مقصورة is pronounced long according to Wehr!
Last edited:
I believe there is. If a word has only one long vowel, the stress is on that syllable. Can you think of any counter-examples?
You can't use that to argue that the rule doesn't exist because the length of that final vowel is being debated here. For me, it's clearly short.
Vowel quality and vowel length are two different things.
Yes, I think كتبنا (we wrote) would be pronounced katabnaa. And أدعو would be ad'uu.
What I mean by quality is that they (the vowels in بنتي) sound completely different (to me). I think this is because one is short and one is long. Exactly like the word mini.
Last edited:
Those examples don't work either! For me, these are "katabna" and "ad3u", with short final vowels. A valid counter-example would be one in which the long vowel was incontrovertibly long.
They sound different to you when said by who? Again, native speakers - at least the ones I've heard - pronounce them the same. If there are any regional or dialectal differences, I'm not aware of them.
Well I can hear it in the song! And I've heard it elsewhere many times. I am not saying the vowels are actually pronounced longer than a short vowel. But they are different sounds. Just like mini.
Are you saying that you hear people pronouncing بنتي exactly as they would pronounce بنتِ ?
Using IPA, to my ears, بنتي is pronounced bɪnti, just as mini is pronounced mɪni. Whereas بنتِ (to my ears) is pronounced bɪntɪ...
I think it might be easier to say that short and long vowels are neutralised at the end of a word in most circumstances. I'm not native, but to my ears if I try and distinguish بنتِ from بنتي, the elongation of the yaa sounds a bit out of place and exaggerated.
I think certain Gulf dialects and probably Iraqi often pronounce final -ii more like -əy or -ɪy which might help distinguish the two.
The final kasra in بنتِ doesn't sound like /ɪ/ to me, it's /bɪnti/. That is to say, the two kasras are pronounced differently in this word because one comes at the end. Iraqi/Kuwait/other dialects might maintain a distinction by way of /bɪntəy/ vs /bɪnti/ but that's probably not a general thing.
I think certain Gulf dialects and probably Iraqi often pronounce final -ii more like -əy or -ɪy which might help distinguish the two.
The final kasra in بنتِ doesn't sound like /ɪ/ to me, it's /bɪnti/. That is to say, the two kasras are pronounced differently in this word because one comes at the end. Iraqi/Kuwait/other dialects might maintain a distinction by way of /bɪntəy/ vs /bɪnti/ but that's probably not a general thing.
Just to mention a similar discussion in another thread.
So would you say the pronunciation of كتبنا (we wrote) and كتبن (they [fem. pl.] wrote) is exactly the same? None of my books would agree with you!
At the risk of sounding like a broken record, yes.
I want to again reiterate that this whole time I've been talking about length and not quality.
AndyRoo, I'm not sure what exactly you're arguing. The quotes below seem contradictory:
Let me add that it doesn't sound bad to slightly lengthen these final vowels, BUT
A.) They will still typically be shorter than a long vowel, such as the dual suffix -aa.
B.) (This is the crucial point!) They don't have to be lengthened (at all) to be pronounced correctly!
In other words, ذهبا has to be pronounced "thahabaa" - with a stressed long vowel. Any other pronunciation is incorrect.
ذهبنا, on the other hand, is typically pronounced "thahabna" - with an unstressed short vowel. That final vowel could be lengthened somewhat - to a length in between "a" and "aa," but it doesn't have to be, and a full-on "aa" would sound wrong.
Finally, my point about length and stress has yet to be countered. I have yet to be presented with an Arabic word that is not of the type being discussed here and has a single long unstressed vowel.
In Czech, vowel length and stress do not covary, so long and short vowels can be stressed or unstressed. I always struggle to pronounce long unstressed vowels in Czech, while the other three types do not pose a challenge to me. This would not be the case if long unstressed vowels were normal or natural in Arabic. If the final vowel in ذهبنا were pronounced long, then I should have no trouble pronouncing the word Česká, for example, where the vowel in the second syllable is long and unstressed. But I always have to force myself to lengthen it when I pronounce the word, and it feels unnatural.
(It would be great if other native speakers posted in this thread, by the way.)
Last edited:
They are not contradictory. I am talking about what I think are two different sounds completely. The actual length is not so important.
I am not sure what you mean by this. May I ask my question again:
Are the vowels in the word بنتي exactly the same? If you extended them both (perhaps if you were shouting the word in the mountains) would the sound be the same for both vowels? I am 99.99% sure they wouldn't be. I think it is just like the word mini, as I said before.
I agree!
I think andyroo is suggesting that even if phonetically these vowels are short, there is some kind of quality difference as well that means that al-binti and bint-ii (transliteration not intended to suggest actual length) are still different in terms of realisation (perhaps [bɪntɪ] vs [bɪnti]?)
I still don't think this is the case, though. I think both final is are pronounced identically - in terms of vowel quality and vowel length. There is a difference between the /i/ in the first syllable (realised as more central or something I think, so maybe [ɪ] is a good phonetic transcription) and the /i/ in the second syllable, but this is because of different allophonic realisations in closed and open syllables.
I still don't think this is the case, though. I think both final is are pronounced identically - in terms of vowel quality and vowel length. There is a difference between the /i/ in the first syllable (realised as more central or something I think, so maybe [ɪ] is a good phonetic transcription) and the /i/ in the second syllable, but this is because of different allophonic realisations in closed and open syllables.
It is important!
Earlier in the thread you were arguing quite clearly that final long-vowel suffixes were actually pronounced long (بلادي بلادي, etc.); in #22, you say there is a difference between a final kasra and a final ي suffix "because one is short and one is long." Now you are talking about quality and saying the length may not actually be different.
Are you conceding that the length is not different? Are you withdrawing your earlier stance and instead abstaining from any judgment on length but insisting that there is a difference in quality?
You say you are not sure what I mean when I say I've been talking about length and not quality. I'm puzzled as to how this is not clear, but let me try to clarify. Quality is about what sound you are actually producing, regardless of duration. Length is about duration. What I've been saying is that the duration of the final vowels in بنتِ and بنتي is the same, or at least both fall within the range of what I would consider "short"; as I said, one could extent the latter somewhat, but not so much that it sounds like the vowel in ريش, for example.
As for quality, I agree with analeeh on all counts.
I think I have been clear in what I said, but to explain again, I will use an English example (as Arabic is in dispute).
In English, there are different pronunciations for the vowel "i". The word "mini" would be written phonemically as /mɪni/ whereas the word "mean" would be written /mi:n/. The /:/ indicates the vowel is lengthened.
The final "i" in "mini" you will note is a shortened version of the (long) vowel in "mean", it is not the same as the first (short) vowel in mini.
This is what I meant when I said it was pronounced long. Maybe I wasn't clear about this to start with, but I explained myself later.
I am not conceding it, I agree! I have never disputed it.
I asked you about the pronunciation of the vowels in the word "بنتي" and you said they were "identical". Now you seem to be shifting and saying they are different. But I am still not sure about your view.
I believe the phonemic pronunciation of بنتي is /bɪnti/ (exactly like mini). Do you agree? Or do you think it should be /bɪntɪ/?
Elroy, are you talking about when the long vowel precedes alif lam, like in OP's example? Because I can't believe that long vowels should never be pronounced at the end of the word. If all endings are pronounced with short vowels except the dual, then I feel like the books I've read have failed to mention something quite significant.
Last edited:
So all this time what you've actually been saying is that both بنتي and البنتِ have the same final vowel, which is pronounced [ i ] and not [ɪ], and using 'long' and 'short' as shorthand for the two allophonic realisations of (quantitatively) short /i/?
(In IPA // is usually used for phonemic transcription and [] for phonetic, by the way).
This may explain the confusion, but it doesn't explain why you think that كتبنا كتبنَ or, even more fundamentally, قميصِ قميصي should be pronounced differently...
... as you asserted here. ^
No, he's talking about general pronunciation. The most elevated prescriptive pronunciation of fuSHaa (the one used for Qur'anic recitation and so on) and presumably the original pronunciation of Classical Arabic may pronounce them as long, but by far the most widespread pronunciation treats them all as short vowels.
It may be that this is not the case for MSA as it is spoken by e.g. Moroccans, who often follow Moroccan rules in stressing final vowels. But my impression of the Moroccan pronunciation of MSA is that it either veers towards what we might call a kind of 'al Jazeera standard' (where final long vowels are short with the exceptions given by elroy above) or else towards spoken Moroccan (where long vowels are not really distinguished from short vowels in the same quantitative sense as they are elsewhere in the Arab world).
I am talking about the two vowels in the word بنتي being different. If we talk about بنتي and البنتِ we add an extra confusion (see next point).
I originally asked about the pronunciation of قميصِ and قميصي to understand if elroy meant the the ي was pronounced the same as a kasra. But I realised there could be a confusion as indeed these two words would be pronounced the same - but not because the ي is pronounced like a kasra (/ɪ/), but because (I think) the kasra changes at the end of the word (to /i/).
That is why I then asked about بنتي.
I think the correct pronunciation of بنتي is /bɪnti/, but you think it is [bɪntɪ].
I think the pronunciation of قميصي (on its own; not followed by الجديد) is: /qami:Si/. I guess you think it is [qami:Sɪ]?
And what is the pronunciation of كتبنا and كتبنَ? Are they both exactly the same and is it /katabna/? It is not my experience.
I said I was using a phonemic transcription. I thought it would be OK here.
A phonemic transcription is not appropriate here for the reasons I outlined above. Arabic has two /i/ phonemes predominantly distinguished by length (if you prefer we can analyse them as one phoneme which can appear stacked twice, but this would be an odd analysis) - /i/ and /i:/. You can write these however you want - /ɪ/ and /i:/, or whatever - the way you write it has no impact on its actual phonetic realisation. If you really wanted you could write it /ɪ/ and /i/, although this would be a misleading choice since the main distinction is vowel length.
I have never claimed (nor has Elroy) that the pronunciation should be [bɪntɪ], nor would this in any case mean that it couldn't be /bɪnti/ (since phonemic analysis and phonetic analysis are different). What we're saying is that final long vowels (with the exception of some final -aa, although not the one in كتبنا, probably because this morpheme - unlike the dual -aa - is not found in colloquials) are neutralised for length. What this means is that the one distinction between /i/ and /i:/ - the length - disappears.
It is true that in some phonological contexts - as in a closed syllable - short /i/ is often realised differently qualitatively as well, as [ɪ]. But this is not the case finally generally speaking (except in perhaps some North Levantine dialects where final -e -i merge or have other weird things going on with them, but I would be surprised to hear this feature carried over into MSA pronunciation). In final position short /i/ is realised as [i ]. Likewise, long /i:/ - which is neutralised for length - is also realised as [i ]. The fact that there may be two different underlying phonemes here (a long /i/ and a short /i:/) is largely irrelevant - phonetically they are realised identically.
There is a rule in basically all eastern dialects that final long vowels cannot be long unless they are stressed (in the Mashriq at least, from Egypt to Iraq, stress and length tend to be very closely associated as non-stressed long vowels are often reduced). This rule is carried over into the usual pronunciation of MSA. Elroy has already said that bintii pronounced long but with the stress on the first syllable sounds and feels weird to him and I agree. That pronunciation would probably be interpreted as a (formal) pronunciation of the nisba ending -iyy (which itself is only pronounced as /ij:/ rather than as /i/ in more formal registers).
To be absolutely clear, I am saying that the pronunciation of both binti and bintii is typically something like [bɪnti]. Depending on how we want to analyse it, we can denote this in phonemic transcription as any of the following:
بنتي بنتِ
/binti binti:/
/binti binti( : )/
/binti binti/
But the differences here reflect differences in analysis and not in pronunciation.
I have never claimed (nor has Elroy) that the pronunciation should be [bɪntɪ], nor would this in any case mean that it couldn't be /bɪnti/ (since phonemic analysis and phonetic analysis are different). What we're saying is that final long vowels (with the exception of some final -aa, although not the one in كتبنا, probably because this morpheme - unlike the dual -aa - is not found in colloquials) are neutralised for length. What this means is that the one distinction between /i/ and /i:/ - the length - disappears.
It is true that in some phonological contexts - as in a closed syllable - short /i/ is often realised differently qualitatively as well, as [ɪ]. But this is not the case finally generally speaking (except in perhaps some North Levantine dialects where final -e -i merge or have other weird things going on with them, but I would be surprised to hear this feature carried over into MSA pronunciation). In final position short /i/ is realised as [i ]. Likewise, long /i:/ - which is neutralised for length - is also realised as [i ]. The fact that there may be two different underlying phonemes here (a long /i/ and a short /i:/) is largely irrelevant - phonetically they are realised identically.
There is a rule in basically all eastern dialects that final long vowels cannot be long unless they are stressed (in the Mashriq at least, from Egypt to Iraq, stress and length tend to be very closely associated as non-stressed long vowels are often reduced). This rule is carried over into the usual pronunciation of MSA. Elroy has already said that bintii pronounced long but with the stress on the first syllable sounds and feels weird to him and I agree. That pronunciation would probably be interpreted as a (formal) pronunciation of the nisba ending -iyy (which itself is only pronounced as /ij:/ rather than as /i/ in more formal registers).
To be absolutely clear, I am saying that the pronunciation of both binti and bintii is typically something like [bɪnti]. Depending on how we want to analyse it, we can denote this in phonemic transcription as any of the following:
بنتي بنتِ
/binti binti:/
/binti binti( : )/
/binti binti/
But the differences here reflect differences in analysis and not in pronunciation.
Last edited:
Just an anecdote: I noticed today talking with Emiratis that
نصفي الثاني was pronounced nə9fiyyi l-thaaniy, not nə9f-il-thaani.
This might only be in Gulf dialects, though.
I imagine قميصي الجديد would be qamii9iyyi l-jadiid in this dialect, which preserves the distinction.
نصفي الثاني was pronounced nə9fiyyi l-thaaniy, not nə9f-il-thaani.
This might only be in Gulf dialects, though.
I imagine قميصي الجديد would be qamii9iyyi l-jadiid in this dialect, which preserves the distinction.
I think part of what is missing in this discussion is what pronunciation is being referred to. In Arabic, there are actual pronunciation rules; however, these pronunciation rules are different than how a native Arabic speaker would speak today.
If we're referring to proper rules, قميصي would be pronounced with a long ee at the end (Qameesee). However, since it is attached to a hamza-tul wasl, in proper pronunciation, the ee would be shortened, and it would sound like Qamees-il-jadeed.
If we're referring to how a native speaker would pronounce it, it would usually be with a shortened 'e' sound (Qameesi). When attached to the next word, it would be Qamis-il-jadeed, just like in proper pronunciation.
If we're referring to proper rules, قميصي would be pronounced with a long ee at the end (Qameesee). However, since it is attached to a hamza-tul wasl, in proper pronunciation, the ee would be shortened, and it would sound like Qamees-il-jadeed.
If we're referring to how a native speaker would pronounce it, it would usually be with a shortened 'e' sound (Qameesi). When attached to the next word, it would be Qamis-il-jadeed, just like in proper pronunciation.
Last edited:
Hi Matat.
Could you clarify please.
What do you think the pronunciation of the "i" is in:
1. Stand alone "Qameesi". I think it is [ i] (= as in mini).
2. "Qamis-il-jadeed". I think it is [ɪ] (= as in mini).
I think everybody is in agreement on this, no?
The example I posted was specifically about certain Gulf dialects which seem to have an alternative -iyy for the 1st person possessive. I don't think this would be standard in MSA or other dialects. I've noticed that -i in general is often -iyy / -ey in the Gulf: anti -> intay, laa truu7i -> laa truu7ey, etc.
It would sound the same as before the wasl, but it would be shortened instead of lengthened. So it would sound more like the second 'i' of mini.
I am not sure! See next point.
I'm sorry Matat, but this is still not clear to me. Are you referring to the "i" in stand alone "Qameesi" here?
Just to reiterate what I said above, there are three different "i" sounds in Arabic at issue here:
1. Short "i" = mid word kasra = phonetic symbol [ɪ]. e.g. the kasra in كِتاب.
2. Long "i" = mid word kasra+yaa' = phonetic symbol [i:]. e.g. the kasra+yaa' in قمِيص.
3. Short long "i" = end of word kasra or end of word kasra+yaa' = phonetic symbol [i ], e.g. the final vowel sound in قميصي and أنتِ.
In formal Arabic, when there is a wasl, the vowel before it is always pronounced short.
e.g. في الباب [fɪlba:b] rhymes exactly with جلباب [jɪlba:b]; and بابي الجديد sounds exactly the same as بابِ الجديد [ba:bɪljaedi:d].
Apologies if this is obvious to everyone. But I just want to make sure we are all agreed on these points.
No, I was referring to when it is a wasl. If it was a standalone, then the ee would be long.
Yes and yes.
No. قميصي, when stood alone, would be pronounced with a long ee at the end. It would be different than أنتِ. With أنتِ, the e sound is shorter.
The e sound at the end of قميصي in the sentence "قميصي جديد" sounds longer than the e sound at the end of أنتِ in the sentence "أنتِ جميلة".
Another difference is that قميصي in pausa would sound no different than when it's not in pausa. So when saying "قميصي" by itself or when saying a continuous sentence like "قميصي جديد", the word قميصي would sound the same with a long ee. For أنتِ, it would be a short e sound in a sentence, but in pausa, you would not pronounce the kasra at all and would sound like أَنْتْ.
The only time the ي in قميصي would sound like the kasra in أنتِ is if it is followed by a hamzatulwasl. So saying قميصي الجديد should sound like you're saying قميصِ الجديد.
Yes.
Not according to elroy!
See my post #6 and the ensuing discussion!
Apologies if I have misunderstood what you are saying
Really?! He should have said!
Yes, I was referring to proper pronunciation rules (not how a native speaker would commonly pronounce it). I said this in my first comment. Sorry if I wasn't clear.