A terrific thunderstorm rages over the mountains on the northern edge of the Najd. The scene is imagined over so vast an area that it must be poetic fiction. (As the medieval commentators note: Sitār, Yaḏbul and Qaṭan cannot possibly all be seen from the same place.)
From the Muˁallaqa: A Mountain Storm
Attributed to Imru' al-Qays
Translated by A.Z. Foreman
Friend, can you see the lightning? There: its flash
bolting like hands in a crownbright cloudheap, quick
to shed light on all things; there: like the lamps
of a hermit who has oiled each coily wick.
I sat to watch it with my friends, between
Ḍārij and Al-ˁUdhayb. Oh I gazed far
enough to see the storm raise its right arm
on Mount Qaṭan, and its left on Al-Sitār,
dumping its rainload hard around Kutayfa
and blowing flat the great Kanahbul trees.
Its shower bucked out over Mount Qanān
panicking all the whitefoot Ibices.
At Taymā' it left not one palm-trunk standing
nor rampart made of anything but rock,
Mount Thabīr in its water-onslaught stood
like a tribe's chieftain in a stripelined cloak.
Come dawn, the upper peaks of Al-Mujaymir
stood spindle-whirled with storm-debris all round,
the flood's bale flung on Al-Ghabīṭ like cloth-sacks
dropped by a Yemeni merchant to the ground.
Come morning, finches noised about the dales
as if blind drunk on pepper-fiery wine.
Come evening, raptors lay drowned at its edge
like squill-roots twisted into a freakish twine.
Audio recording in reconstructed Early Classical Arabic pronunciation
A more rhythmicized recitation
The Arabic of my audio recording is in what I choose to call "Early Classical Arabic" pronunciation. To put it grandiosely, it is a kind of Arabic that hasn't been heard for over a thousand years. To put it plainly, it is a speculative reconstruction of the kind of Arabic pronunciation the grammarian Sibawayh might have used, based on his description of Arabic speech-sounds, and augmented with some inference based on typology. The main differences from textbook Classical Arabic as it is taught and learned today are as follows:
The ج was /ɟ/ (not /dʒ/)
The ش was /ɕ/ (not /ʃ/)
The ص was (for an indeterminate number of speakers including Sibawayh) an affricate /t͡sˤ/ rather than the fricative /sˁ/. (For reasoning behind this reconstruction see this article by Ahmad Al-Jallad).
The ض was a pharyngealized lateral, probably /ɮˤ/ or /d͡ɮˤ/ (the modern /dˁ/ pronunciation is much more recent)
The ت and ك appear to have been quite strongly aspirated /kʰ tʰ/.
In addition to the familiar three vowels /a: i: u:/ there existed /e:/ for many speakers (and, more marginally, /o:/ for some.)
The vowel /a:/ was optionally raised to [e:] due to i-mutation under a complex of different circumstances, partially neutralizing the contrast between /a:/ and phonemic /e:/ and giving the realizations of /a:/ a range and distribution not commonly heard in modern elevated poetic recitations.
Although I render ش as alveolo-palatal /ɕ/, full disclosure requires noting that another possibility would be a true palatal non-sibilant /ç/, which is what many (perhaps most) posit based on a strict interpretation of Sibawayhi's statement. Now, Sibawayhi, who doesn't get enough credit as a phonetician, could probably have distinguished palatal from alveolo-palatal articulation. But whether he would have cared to is a different question. Although he groups ي ش ج at the same place of articulation, it is only ش which causes assimilation of the definite article. Thus there was something about šīn that made it pattern, for assimilation purposes, with the coronals rather than the dorsals. The most straightforward interpretation would be that this is because šīn was indeed a sibilant. Sibilants as an articulatory class involve a centerline grooved tongue focusing the airstream such that it strikes the teeth. Whereas non-sibilant fricatives do not involve the teeth as a secondary articulator. Sibilants, probably because of the need to involve the teeth, are always coronal. Alveolo-palatal articulation sits uneasily in a no-man's land between the dorsal and coronals, and is as far back as you can go and still produce a sound that behaves acoustically and structurally like a sibilant. For /ç/ to function as a sibilant, it must thus have front articulation [ç̟], which (notational and theoretical games aside) makes it functionally /ɕ/.
One phonologically interesting way in which Sibawayhi's Arabic was likely counterintuitive from the standpoint of many modern accents of the standard language, and doubly so for non-native Arabic speakers given how they tend to be taught, is that what we normally think of as voiced plain stops /b d ɟ/ and voiceless plain stops /t k/ did not — strictly speaking — have presence or absence of voicing per se as their distinguishing feature. In this, Sibawayh's Arabic would align with certain modern dialects like San'ani Arabic. (See Phonation and glottal states in Modern South Arabian and San’ani Arabic by Janet Watson and Barry Heselwood for this and more, including a good explanation of a crucial articulatory category in Sibawayh's description.) The chief featural distinction between the two sets was probably aspiration in the latter and non-aspiration (with adductive glottal tension) in the former. In a dialect like this, although /b d ɟ/ probably did not have fully specified voicing, much of the time this would be of little phonetic consequence since in most positions voicing would be triggered positionally. In post-pausal position, however, although /b d ɟ/ would trigger a glottal prephonation state, their actual voice-onset time would not necessarily be different from that of a voiceless non-aspirated stop.
The Original:
From the Muˁallaqa: A Mountain Storm
Attributed to Imru' al-Qays
Translated by A.Z. Foreman
Friend, can you see the lightning? There: its flash
bolting like hands in a crownbright cloudheap, quick
to shed light on all things; there: like the lamps
of a hermit who has oiled each coily wick.
I sat to watch it with my friends, between
Ḍārij and Al-ˁUdhayb. Oh I gazed far
enough to see the storm raise its right arm
on Mount Qaṭan, and its left on Al-Sitār,
dumping its rainload hard around Kutayfa
and blowing flat the great Kanahbul trees.
Its shower bucked out over Mount Qanān
panicking all the whitefoot Ibices.
At Taymā' it left not one palm-trunk standing
nor rampart made of anything but rock,
Mount Thabīr in its water-onslaught stood
like a tribe's chieftain in a stripelined cloak.
Come dawn, the upper peaks of Al-Mujaymir
stood spindle-whirled with storm-debris all round,
the flood's bale flung on Al-Ghabīṭ like cloth-sacks
dropped by a Yemeni merchant to the ground.
Come morning, finches noised about the dales
as if blind drunk on pepper-fiery wine.
Come evening, raptors lay drowned at its edge
like squill-roots twisted into a freakish twine.
Audio recording in reconstructed Early Classical Arabic pronunciation
A more rhythmicized recitation
The Arabic of my audio recording is in what I choose to call "Early Classical Arabic" pronunciation. To put it grandiosely, it is a kind of Arabic that hasn't been heard for over a thousand years. To put it plainly, it is a speculative reconstruction of the kind of Arabic pronunciation the grammarian Sibawayh might have used, based on his description of Arabic speech-sounds, and augmented with some inference based on typology. The main differences from textbook Classical Arabic as it is taught and learned today are as follows:
The ج was /ɟ/ (not /dʒ/)
The ش was /ɕ/ (not /ʃ/)
The ص was (for an indeterminate number of speakers including Sibawayh) an affricate /t͡sˤ/ rather than the fricative /sˁ/. (For reasoning behind this reconstruction see this article by Ahmad Al-Jallad).
The ض was a pharyngealized lateral, probably /ɮˤ/ or /d͡ɮˤ/ (the modern /dˁ/ pronunciation is much more recent)
The ت and ك appear to have been quite strongly aspirated /kʰ tʰ/.
In addition to the familiar three vowels /a: i: u:/ there existed /e:/ for many speakers (and, more marginally, /o:/ for some.)
The vowel /a:/ was optionally raised to [e:] due to i-mutation under a complex of different circumstances, partially neutralizing the contrast between /a:/ and phonemic /e:/ and giving the realizations of /a:/ a range and distribution not commonly heard in modern elevated poetic recitations.
Although I render ش as alveolo-palatal /ɕ/, full disclosure requires noting that another possibility would be a true palatal non-sibilant /ç/, which is what many (perhaps most) posit based on a strict interpretation of Sibawayhi's statement. Now, Sibawayhi, who doesn't get enough credit as a phonetician, could probably have distinguished palatal from alveolo-palatal articulation. But whether he would have cared to is a different question. Although he groups ي ش ج at the same place of articulation, it is only ش which causes assimilation of the definite article. Thus there was something about šīn that made it pattern, for assimilation purposes, with the coronals rather than the dorsals. The most straightforward interpretation would be that this is because šīn was indeed a sibilant. Sibilants as an articulatory class involve a centerline grooved tongue focusing the airstream such that it strikes the teeth. Whereas non-sibilant fricatives do not involve the teeth as a secondary articulator. Sibilants, probably because of the need to involve the teeth, are always coronal. Alveolo-palatal articulation sits uneasily in a no-man's land between the dorsal and coronals, and is as far back as you can go and still produce a sound that behaves acoustically and structurally like a sibilant. For /ç/ to function as a sibilant, it must thus have front articulation [ç̟], which (notational and theoretical games aside) makes it functionally /ɕ/.
One phonologically interesting way in which Sibawayhi's Arabic was likely counterintuitive from the standpoint of many modern accents of the standard language, and doubly so for non-native Arabic speakers given how they tend to be taught, is that what we normally think of as voiced plain stops /b d ɟ/ and voiceless plain stops /t k/ did not — strictly speaking — have presence or absence of voicing per se as their distinguishing feature. In this, Sibawayh's Arabic would align with certain modern dialects like San'ani Arabic. (See Phonation and glottal states in Modern South Arabian and San’ani Arabic by Janet Watson and Barry Heselwood for this and more, including a good explanation of a crucial articulatory category in Sibawayh's description.) The chief featural distinction between the two sets was probably aspiration in the latter and non-aspiration (with adductive glottal tension) in the former. In a dialect like this, although /b d ɟ/ probably did not have fully specified voicing, much of the time this would be of little phonetic consequence since in most positions voicing would be triggered positionally. In post-pausal position, however, although /b d ɟ/ would trigger a glottal prephonation state, their actual voice-onset time would not necessarily be different from that of a voiceless non-aspirated stop.
The Original:
أصَاحِ تَرَِى بَرْقاً أُرِيْكَ وَمِيضَـهُ
كَلَمْـعِ اليَدَيْنِ فِي حَبِيٍّ مُكَلَّـلِ
يُضِيءُ سَنَاهُ أَوْ مَصَابِيْحُ رَاهِـبٍ
أهَانَ السَّلِيْـطَ بِالذُّبَالِ المُفَتَّـلِ
قَعَدْتُ لَهُ وصُحْبَتِي بَيْنَ ضَـارِجٍ
وبَيْنَ العـُذَيْبِ بُعْدَمَا مُتَأَمَّـلِ
عَلَى قَطَنٍ بِالشَّيْمِ أَيْمَنُ صَوْبِـهِ
وَأَيْسَـرُهُ عَلَى السِّتَارِ فَيَذْبُـلِ
فَأَضْحَى يَسُحُّ المَاءَ حَوْلَ كُتَيْفَةٍ
يَكُبُّ عَلَى الأذْقَانِ دَوْحَ الكَنَهْبَلِ
ومَـرَّ عَلَى القَنَـانِ مِنْ نَفَيَانِـهِ
فَأَنْزَلَ مِنْهُ العُصْمَ مِنْ كُلِّ مَنْـزِلِ
وتَيْمَاءَ لَمْ يَتْرُكْ بِهَا جِذْعَ نَخْلَـةٍ
وَلاَ أٌجُماً إِلاَّ مَشِيْداً بِجِنْـدَلِ
كَأَنَّ ثَبِيْـراً فِي عَرَانِيْـنِ وَبْلِـهِ
كَبِيْـرُ أُنَاسٍ فِي بِجَـادٍ مُزَمَّـلِ
كَأَنَّ ذُرَى رَأْسِ المُجَيْمِرِ غُـدْوَةً
مِنَ السَّيْلِ وَالأَغثَاءِ فَلْكَةُ مِغْـزَلِ
وأَلْقَى بِصَحْـرَاءِ الغَبيْطِ بَعَاعَـهُ
نُزُوْلَ اليَمَانِي ذِي العِيَابِ المُحَمَّلِ
كَأَنَّ مَكَـاكِيَّ الجِـوَاءِ غُدَّبَـةً
صُبِحْنَ سُلافاً مِنْ رَحيقٍ مُفَلْفَـلِ
كَأَنَّ السِّبَـاعَ فِيْهِ غَرْقَى عَشِيَّـةً
بِأَرْجَائِهِ القُصْوَى أَنَابِيْشُ عُنْصُـلِ
Posts
