Arabic Quantitative Metrics in al-Zamakhshari’s al-Qisṭâs al-Mustaqîm A Senior Honors Thesis Presented in Partial Fulfillment of the Requirements for Graduation with research distinction in Arabic in the Undergraduate colleges of the Ohio State University by Leopold Lewis Eisenlohr The Ohio State University March 2011 Project Advisor: Professor Bruce Fudge, Near Eastern Languages and Cultures 2 Arabic Quantitative Metrics in al-Zamakhshari’s al-Qisṭâs al-Mustaqîm Leopold Eisenlohr Contents Introduction to the Construction of Arabic Poetry..............................................................5 An Approach Toward Determining Meter and Variation through al-Qisṭâs al-Mustaqîm............................................................................9 Text of al-Qisṭâs al-Mustaqîm...........................................................................................30 Glossary of Poetic Terminology......................................................................................101 Changes Permissible by Foot...........................................................................................105 Glossary of al-Rabaˁî.......................................................................................................111 Xalîlian Metrics in Relation to the Oral Inheritane of Arabic Poetry..............................114 3 A Note on the Editions Used The main edition of al-qisṭâs al-mustaqîm used in this study is that edited and introduced by Dr. Bahijah Baqir al-Hasani, publoished in Baghdad by the Maktabah al- Andalus in 1969. Other editions consulted include an edition by Dr. Fakhr al-Din Qabawah published in 2008 in Damascus by Dâr Hârûn al-Raṣîd; an electronic version in the cd-rom al-mawsûˁa(t) al-šiˁriya(t) published in 2003 in Abu Dhabi by al-Majmaˁ al- Thaqafî; as well as an 1853 manuscript copy accesed online through the King Saud University Library. These editions all differ slightly from one another, and in those places the effort has been made to provide the work in its most complete form possible; therefore verses omitted from the edition of Dr. Bahijah al-Hasani but included in that of Dr. Qabawah have been included in the translation. In places where typographical errors appear to have been made, alternate versions have been sought of the line in question and those conforming most closely with the specifications of variation have been used. Other medieval Arabic works on ˁarûḍ which were read in comparison with that of al-Zamakhsharî are the kitâb al-ˁarûḍ of Ibn Jinnî, the kitâb al-ˁarûḍ of al-Akhfash, the ˁarûḍ al-waraqah of al-Jawharî, the kitâb al-ˁarûḍ of al-Rabaˁî al-Naḥwî, and the kitâb al-kâfî fî al-ˁarûḍ wa al-qawâfî of al-Khaṭîb al-Tibrîzî. These works have not been cited in the text, but citations will be given for the editions used. 4 Jâr Allâh al-Zamaxšarî is known for his grammatical and theological works, including his important tafsîr. Other important works are his mufaṣṣal on grammar, in which can be seen the high degree of influence on him by Sibawayhi’s kitâb; his extensive dîwân; a commentary on the lâmiyya(t) al-ˁarab; and a book on syntax called kitâb al-mufrad wa-l-muˀallaf fî al-naḥw. He was an adherent of the theological school of Muˁtazilism, which was introdiced into his native land by his teacher Abu Mudâr Maḥmûd ibn Jarîr al-Iṣbahânî; this can be seen in his Quranic commentary but is not readily apparent in his grammatical writings. He did not subscribe to the Persian šuˁûbiyya(t) but was a proponent of Arab culture and believed in the superiority and divinity of the Arabic language. It is this fact of his life that most heavily influences his approach to ˁarûḍ in his loyal adherence to classical Arabic traditions. 5 Introduction to the Construction of Arabic Poetry These are the characters used for transliteration, in order of the Arabic alphabet, with the IPA equivalents given: â/a/- b t ƿ(θ) j(dʒ) ḥ(ħ) x d ð r z s š(ʃ) ṣ(sˤ) ḍ(dˤ) ṭ(tˤ) ẓ(zˤ) ˁ(ʕ) ġ(ɣ) f q k l m n h w/û/u y/î/i Hamzah (glottal stop) = ˀ(ʔ) The feminine tâˀ ending is written in pause with a ‘t’ in parentheses so hamza(t). Assimilated consonants are written as pronounced, so ˀad-dâˀiratu -ƿ-ƿâniya. Final long vowels before initial joining Alef (al-waṣl) are shortened and joined to the following consonant, so fi -ð-ðuˁri (رعذلا يف). Final kasra(t) and ḍamma(t) are often elongated in pronunciation into their long forms, which is represented in transliteration but not in the Arabic text, so wa-sayfihî wa-rumḥihî wa-yaḥtamî (يمتحيو هحمرو هفيسو). Arabic prosody is based around feet and counted syllables. There are short and long vowels: a and â, i and î, u and û. The symbol “/” represents a consonant with a short vowel, as the ki in kitâb, while the symbol “o” represents either a) an elongation of the vowel (i.e. changing a short a, i, or u - the three vowels in Arabic - into long â, î, and û) or b) a vowelless consonant after a vowel, which is grouped with its preceding consonant-short vowel pair. The first case is represented by the kâ in kâtib and the second by tak in taktubu. Two “o” symbols in a row indicate a superlong syllable, which is the combination of a long vowel with a vowelless consonant at the end, as in the second syllable of kitâb. Therefore these words would be represented: kitâb //oo (/ = ki; / = ta; o = lengthening the short a; o = vowelless b) 6 kâtib /o/o (/ = ka; o = long a; / = ti; o = vowelless b) taktubu /o// (/ = ta; o = k; / = tu; / = bu) The symbol “/” is called mutaḥarrik “movent” and the symbol “o” is called sâkin “quiescent.” These terms are important in understanding the structure of a meter and its exceptions, as works on prosody frequently give instructions like “delete the fifth sâkin,” which would be unintelligible if using another symbolic system that does not use the Arabic grammarians’ symbols for mutaḥarrik and sâkin. They are also used to distinguish between feet that are composed of five and seven elements, which are the “/” and “o” marks. This method differs from symbolic representations in Western poetics, but is preferable as it is easily understood and allows for a modal (i.e. working within the system) understanding of the roles of the syllable components. These signs are used to quantify meters, and are the building blocks of the feet. There are eight feet which in combination form the meters. They are arranged in circles, which are the Muxtalif, Muˀtalif, Muštabih, Mujtalab, and Muttafiq. The rhythmic patterns of meters in the same circle are basically the same, only the meters begin on different positions of the same repearing pattern, like a drum beat repeated starting on the second quarter note, or the third, or the fourth, in relation to the basic beat beginning on the first: in repetition they encompass the same rhythmic flow but given different accent. One line is actually composed of two halves (each one called a hemistich or half- line) which mirror each other in the feet they are composed of (but not in the metrical variations that can occur in them). Sometimes a meter can occur as what is called a 7 maṣraˁ, or a lone hemistich acting as the full line. The parts of the line (or bayt) have names which one must know when determining the metrical variations. In the first hemistich, the last foot is called the ˁarûḍ. In the second hemistich, the last foot is called the ḍarb. The middle part in each hemistich is called the ḥašw. The eight feet are faˁûlun, fâˁilun, mustafˁilun, mafâˁîlun, fâˁilâtun, mafâˁalatun, mutafâˁilun, and mafˁûlâtu. These are artificial names created to represent the syllabic quantities in a meter, and are important in understanding the workings within a line. So, for example, the meter Tawîl is said to be on the pattern of faˁûlun mafâˁîlun faˁûlun mafâˁîlun - faˁûlun mafâˁîlun faˁûlun mafâˁîlun. However, it must be kept in mind that it is actually on the pattern of //o/o,//o/o/o,//o/o,//o/o/o - //o/o,//o/o/o,//o/o,//o/o/o, or a fourfold repetition of “short long long, short long long long.” Each meter has certain irregularities which are permissible and which are linked to the feet. In the following translation of al-Zamaxšarî’s work the characteristics of each meter and foot are laid out in what purports to be an exhaustive catalog, though through comparison with other works on ˁarûḍ (the Arabic term for this science of the poetic meters) in Arabic and English - and indeed with other versions of the same work - it is apparent that there does not exist a monolithic guide to the meters and their exceptions. In reality there are no great differences in the overall metrical theory; only, in the galaxy of regulations imposed as to where a sâkin or a mutaḥarrak can be legally elided, modified, or added, complete agreement between scholars would apparently be hard to accomplish. The concept of the circles from which the meters derived is important in relating the meters to each other. There are five circles, each of which is a loop of mutaḥarrik-s and sâkin-s arranged so that by beginning in one place one meter will be found, while 8 beginning a spot over will generate another meter. The meters in the same circles are therefore “shifted over” versions of each other; that is, in the pattern [=>//o/o//o/o/o//o/o//o/o/o=>] which is circular, which base rhythm of the circle known as Muxtalif, the three other meters which are part of the circle can be derived from starting at different points. Here is where the meters of Muxtalif begin. Starting from the beginning yields the Tawîl. / /o /o / /o /o /o / /o /o / /o /o /o fa ˁû lun ma fâ ˁî lun fa ˁû lun ma fâ ˁî lun ③ ② ① Beginning at ① yields the Madîd; at ②, the Basîṭ; at ③, the so-called Muhmal, though this in al-Zamaxšarî is not used except in a modified form as the Muqtaḍab (mufˁalâtu muftaˁilun) which is actually considered as part of the Muštabih circle. This is, however, thoretical. It will be noticed that this is the exhaustion of this circle - if it were begun still one place earlier it would once again simply be the Tawîl. Finally, Arabic poems mostly exhibit rhyme on the end of a line (bayt) such that, for example, each verse would end in the sound lî. Rhyme, or qâfiya(t) in Arabic, has been the subject of a number of scholarly works but is not discussed in this study, which is limited to meter. Qaṣîda(t)-s are often named for their rhyme, so qaṣîda(t) bâˀiyya(t) = qaṣîda(t) with rhyme on the letter bâˀ. 9 An Approach toward Determining Meter and Variation Based on al-Qisṭâs al-Mustaqîm Understanding Arabic prosody in Western poetic terminology obscures the workings of the metrical systems and for this reason the Arabic system of ḥarf transcrption will be maintained throughout this study - i.e. the representation of a mutaḥarrik with ‘/’ and a sâkin with ‘o.’ There is no reason not to let the Arab prosodists be our teachers instead of converting their system into one of our own necessarily forced construction, and so this study will use little English terminology. This will necessitate the heavy use of glossaries, tables, and lists, but I think it assists in keeping the system as simple as it can be. To this is added the factor of the modality of many Arabic poetic terms, which act within the system and therefore require complex directions instead of direct glosses to Western approximates. Supporting this reasoning is the differentiation between Western and Arabic, and to an extent Persian and Turkish which later adapted to the prestige poetic model of Xalîlian ˁarûḍ (Johanson 8), basis of rhythm - the Western accentual, the Arabic quantitative. There are benefits of being able to determine the meter of a verse. The accuracy with which the poetry adheres to the prescribed meters and catalexes can give an indication of the provenance of an edition, in that if meter is satisfied the author has more credibility elsewhere in the work. Were an editor to copy down his memory of a verse or 10 try reconstructing it on a non-existent (on non-Xalilian) meter, this carelessness may creep like a zaḥâf into the rest of the work. Knowing the meter and its exact exceptions is also useful in analytical approaches to classical Arabic poetry and has implications as to the orality or reconstructedness of pre-Xalilian poetry. Not least is the aesthetic appreciation of the poetic art, which has probably been elusive to Western scholars. While it is one thing to understand the rules of prosody, it is another to understand why they were so important - why the Arabs and Arabic-speaking Muslims were so obsessive about the exhaustiveness of their works on prosody. There are two vantages to the view of a verse: knowing the meter and wanting to arrive at the division of its feet and the acting irregularities; and not knowing the meter with the added question of whether or not the vowelling, and from there the symbolic notation, can be precisely defined. The second case is the more daunting (and much more common), because even if the vowelling is known with certainty it can still produce a situation as in Manzilatun ṣamma ṣadâhâ wa-ˁafat ˀarsumuhâ ˀin suˀilat lam tujibî /o///o/o///o/o///o /o///o/o///o/o///o the muzâḥaf hexameter version of the Kâmil exhibiting the irregularity known as xazl, which is difficult to determine on the basis of the symbols (/,o) alone. How is one to know whether the first foot is fâˁilatun, fâˁilu, mustaˁilun, []fâˁalatun with the first ḥarf+short vowel elided, or, what it actually is according to the system, muftaˁilun? There are certain traits that can, at least, assure the reader that the line is written in accordance with Xalilian prosody.