History of the Oud

History of the Oud

1. The Term ‘ud

Detail from Maghrebi miniature, 13th cent. Literally, ‘ud means ‘twig’, ‘flexible rod’ or ‘aromatic stick’, and by inference ‘piece of wood’. In Ibn Khaldun (14th century), ‘ud denoted the plectrum of the lute called barbat. The etymology of the word has occasioned numerous commentaries, among them Farmer’s alluring thesis that the Arabs adopted the term to differentiate the instrument, with its wooden sound-table, from the similar Persian barbat, whose belly is covered with skin. But this can no longer be defended. The choice of the term ‘ud depends on a discursive form of Arab thought which required some other word to define the barbat before the ‘ud (the same applies to all the instruments of the emergent Islamic world): in this system of ideas, one term refers beck to another or is glossed by yet another, leading to a multiplicity of terms. As the sanj is described as a wanj, the buq as a qarn, the duff as a tar, the ‘ud becomes a synonym of the barbat. The skin-wood difference was not taken into account. This play of reference is clear as stated by the 10th-century Andalusian writer, Ibn ‘Abd Rabbihi: ‘the ‘ud is the barbat‘. Other writers, such as Ibn Sina and Ibn Khaldun, included the ‘ud under the heading of ‘barbat‘ when speaking of its characteristics. In the 10th century, commentaries on pre-Islamic poetry by al-Anbari (d 916) give the ‘ud two semantic meanings: barbat and mizhar; mizhar was to become a poetic substitute for the ‘ud. Earlier, it could equally denote the lyre, suggesting a process of transference from lyre to lute, the lute gradually acquiring the attributes of previous string instruments and becoming a sublimation of them. This transference is noticeable in the earliest Arabic versions of the Bible, where kinnor (lyre) is translated as ‘ud (lute).
2. Early History

The transfer of terms for lyre and lute appears more subtly in the myth of the invention of the ‘ud which has been handed down in two variants from the 9th and 10th centuries, the first being Iraqi and the second Iranian. These say that the ‘ud was invented by Lamak, a direct descendant of Cain; on the death of Lamak’s son, he hung his remains in a tree, and the desiccated skeleton suggested the form of the ‘ud (a contradiction between archaeological research and mythological tradition; the former assumes a process of evolution from lyre to lute, confirmed by organology). The myth attributes the invention of the mi’zaf (lyre) to Lamak’s daughter. There is a Chinese legend that the pipa lute was created through the division or modification of earlier instruments of the zither type, zheng and zhu; such a relationship, of lyre to lute or zither to lute, did exist in the Far East. The comparison suggests a common origin for the ‘ud and the pipa.

Just as the ‘ud becomes the quintessence of earlier chordophones, it also constitutes their functional synthesis. In the 9th century, Miwardi, the jurist of Baghdad, extolled its use in treating illness, a principle allowed and defended in Arab Spain by the 11th-century theologian Ibn Hazm. The symbolism lived on until the 19th century: ‘the ‘ud invigorates the body. It places the temperament in equilibrium. It is a remedy… It calms and revives hearts’ (Muhammad Shihab al-Din). There is also evidence that it was played on the battlefield. In any case it was predominantly in secular usage that the ‘ud made its mark, as the only kind of accompaniment to a form of responsorial song known as sawt, according to written tradition (the Kitab al-Aghani of al-Isfahani) and oral tradition (Tunisia and the Arabian Gulf).

The emergence of the ‘ud on the stage of history is an equally complex matter. Two authors of the end of the 14th century (Abu al-Fida, or Abulfedae, and Abu al-Walid ibn Shihnah) place it in the reign of the Sassanid King Shapur I (241-72). Ibn Shihnah added that the development of the ‘ud was linked to the spread of Manicheism, and its invention to Manes himself, a plausible theory because the disciples of Manes encouraged musical accompaniments to their religious offices. Reaching China, their apostolate left traces of relations between West and East, seen in a short-necked lute similar to the ‘ud. But the movement’s centre was in southern Iraq, whence the ‘ud was to spread towards the Arabian peninsula in the 7th century. However, the texts mentioning the introduction to Mecca of the short-necked lute as the ‘ud were all written in the 9th and 10th centuries. The ‘ud spread to the West by way of Andalusia.

3. Description

Detail from Andalucian miniature, 13th cent. The ‘ud consists of a large soundbox connected to a short neck, features that give it its letters patent of nobility and distinguish it from the long-necked lute family (tanbur, saz, baglama, setar etc). The body has evolved considerably from the original pear shape (which is perpetuated in our own time by the qanbus, taking on a swelling, rounded form). A spherical shape may even have been envisaged: al-Kindi (9th century) described the body of the lute as a ball divided in two, but a century later the Ikhwan al-Safa’ encyclopedia suggested harmonious proportions: ‘The length must be one and a half times the width; the depth, half the width; the neck, one quarter of the length’. If the neck measured only 20 cm (its approximate length today), the total length would be 80 cm, with or without the pegbox – an instrument much the same size as very large contemporary models. Another tradition required the length of the vibrating string from nut to bridge, now about 60 cm, to be equal to the body length, which would leave only 15 cm for the length of the neck.

The body is made from lightweight wood. It consists of a series of 16 to 21 ribs, mentioned as early as the 10th century by the name of alwah (‘boards’) and now called dulu (‘sides’). In the 19th century, the body was called qas’a (‘receptacle’, ‘bowl’), and by the classical authors jism (‘body’). It consists of a strongly rounded back (zahr) and a flat front surface (batn: ‘belly’; sadr: ‘chest’, or wajh: ‘face’) made of lightweight wood, which must ‘reverberate if it is struck’ (Ikhwan al-Safa’). This, the soundboard, is pierced by one quite large soundhole, or (earlier) two small ones; sometimes there are three round or oval soundholes (a design inspired by the lotus flower in Morocco). The holes may be plain or richly ornamented. They are called shamsiyya (‘little sun’), qamarat (‘moons’) or ‘uyun (‘eyes’). The bridge, on the lower part of the belly, is known in classical writings as musht (‘comb’) and as faras (‘horse’) or marbat (‘fastening place’) today. It bears the strings and stands about 10cm from the lower edge, which is called ka’b (‘heel’). (In a recent innovation by Munir Bashir, of Iraq, the 11th, low string is not on the traditional bridge but on the lower edge of the soundbox. The raqma (‘membrane’), a piece of fish-skin or leather, or occasionally of shell, between the bridge and the soundhole, protects the belly from the strokes of the plectrum. This section may take all kinds of extravagant shapes; a Tunisian example is in the form of a parallelogram. The raqma tends to be absent from the modern Iraqi ‘ud.

The neck, joined to the body, is described as ‘unq (‘neck’) in classical writings and the raqba (‘neck’) or zand (‘wrist’) today. It extends the upper part of the instrument by some 20 cm and is inserted into the soundbox up to the soundhole. This length, which has been much discussed, is important in the instrument’s construction, determining the number and location of the intervals and thus affecting the modes. In early 19th-century Egypt, Villoteau gave the measurement as 22.4 cm; a century later, also in Egypt, Kamil al-Khula’i gave it as 19.5 cm. In contemporary Egypt, the length of the neck may vary between 18 and 20.5 can. It is standardized as 20 cm in Syria, but a length of 24.5 cm may be found on Moroccan models, he ‘ud ‘arbi (Arab ‘ud). If the ‘ud ‘arbi is the descendant of an archaic model of Andalusian provenance, the upper part of the instrument may have become shorter. The neck rarely has frets (dasatin), but some are found on the Tunisian lute of Khumayyis Tarnan (1894-1964). Both sides of the neck are inlaid with marquetry to facilitate the learning of the instrument, so providing visual references for the placing of the hand. There is a nut of ivory or bone, called anf (‘nose’) or ‘ataba (‘threshold’), at the upper end of the neck before it bends sharply back to become the pegbox. The tuning-pegs are screwed to the pegbox; they are called mafatih (‘keys’) or more commonly malawi (‘folds’, ‘whorls’). The vibrating length of the strings ranges from 60 to 67cm, according to the model, but lengths as small as 52 cm have been noted.

The quality of material used in the making of the ‘ud is extremely varied; the more the diversity, the better it sounds. This explains the elaborate attention paid to decorative inlay work and the assembling of an impressive number of pieces of wood. The Baghdad lute maker Hanna Hajji al-‘Awwad (1862-1942) used 18,325 pieces to make a single ‘ud.

Classical lexicographers regarded the wood of the wa’s, which cannot be identified, as best for the material of the ‘ud. All kinds of wood have been used, some chosen for their aromatic quality (like sandalwood). Some texts recommend the use of a single type (Ibn Tahhan, 14th century); woods mentioned include walnut, larch, beech, maple, cypress, pistachio, oak, mahogany, cedar and pine for the belly, and ebony for the fingerboard. There is a growing tendency to add inlay work to the ‘ud, whose weight may exceed 800 grams in Arabian lutes but is less in Turkish ones (which are 6 to 8 cm smaller than their Arabian counterparts, and more like the Maghribi ‘ud of the ‘arbi type).
4. Models of the ‘ud

(i) Two-string ‘ud:The thesis of its existence has been upheld by musicologists from Europe and Iran; it envisages the archaic ‘ud as a counterpart of the tanbur, having two strings like that instrument. The argument rests on the names of the strings, two of which are Iranian terms (bamm and zir) and two others of Arab origin (mathna and mathlath). There is no circumstantial documentary evidence to support this hypothesis.

(ii) Four-course ‘ud: The Arabian ‘ud qadim (ancient lute), in particular, invited cosmological speculation, linking the strings with the humours, the temperature, the elements, the seasons, the cardinal points, the zodiac and the stars. The strings may be tuned bass to treble or treble to bass. Bass to treble tuning is represented by al-Kindi (9th century), who advocated tuning the lowest course (bamm or first string) to the lowest singable pitch. Placing the ring finger on a mathematically determined length of this string, one moves on to deduce the pitch of the third open course (mathna), then that of the second (mathlath) and finally the fourth (zir). (This system is also applied to the five-course ‘ud and is still used as a tuning method, following the sequence 1-4-2-3-5 or 1-4-2-5-3.) Adherents of the opposite school (Ikhwan al-Safa’) tune from treble to bass. The intention, inherited in part by the Turkish ‘ud, entails pulling hard on the zir (high) string, so that as it approaches breaking-point it gives a clear sound. One then moves on to determine the pitch of the second course (mathna), the third (mathlath) and finally the fourth (bamm). These two schools did not remain entirely separate. But whichever procedure is used, both end up with tuning by successive 4ths, each course being tuned a 4th above the lower course preceding it. Musicologists, Eastern as well as Western, who try to interpret the pitch of these notes in European terms end up with different results.

Although the four-course ‘ud survives in Morocco, as the ‘ud ‘arbi, the tuning does not conform to the pitches inferred from classical treatises: a conflict between oral and written traditions. The Moroccan method seems to be the product of a previous system, the ‘ud ramal, which also comprised a sequence of 4ths: ramal (?e), hsin, (?a), maya (?d’), raghul (?g’). This ‘ud, like its Tunisian counterpart, may be variously tuned: a feature of these tunings is that they juxtapose the traditional 4ths with the octave and sometimes the 5th and 6th (D-d-G-c). The strings of the ‘ud ‘arbi are named dhil, ramal, maya, hsin; this terminology by no means refers to a fixed pitch standard such as academic and standardized tuition methods would wish for.

At the time of al-Kindi, two of the courses were made of gut and two of silk. In the 10th century silk became predominant and some texts give the composition of the twisted threads: bamm = 64 threads, mathlath = 48, mathna = 36, zir = 27. The figures for the lower courses of the ‘ud correspond with those of two upper strings of the Chinese qin, a fact that has led to speculation about the relationship between Arab and Chinese civilizations by way of the Silk Route.

Another characteristic of the four-course ‘ud is that it is bichordal, having double courses. 13th-century iconography shows that it was already usual to pair the strings at that time, probably to increase sonority but also to allow the development of a more virtuoso type of performance.

(iii) Five-course ‘ud: The addition in Andalusia of a fifth course has been attributed to Ziryab (8th-9th century), although in theoretical writings it appeared in Iraq with al-Kindi. (The addition of this extra course has a parallel in China.) With Ziryab the fifth course, known as awsat (‘intermediary’), a term perpetuated in the ‘ud of San’a’ called qanbus, is placed between the second (mathna) and third (mathlath) courses. With al-Kindi and his successors, it was to reach the end of the instrument and become the string called hadd (‘high’) or the second zir. (According to oral tradition, to obtain an octave on the long-necked lute baglama, a low string should be placed in the middle. This is done when the neck has few frets.) As the ancient ‘ud did not have a two-octave compass, the appearance of the fifth string corresponded to the demands of a new system. The four-course ‘ud had no need to run right through the octave. Its repertory was performed on a tetrachord or pentachord, transposable an octave higher. With the five-course model, the heptatonic system imposed complete series of octaves. The new lute was called ‘ud kamil (‘perfect ‘ud‘).

The five-course ‘ud is the most common and most popular model among performers. It has also been called the ‘ud misri (Egyptian) because of the finely constructed instruments produced by the lute makers of Egypt, who export them as far as Zanzibar. The people of North Africa have added the dialectal name of m’sharqi or mashriqi (‘of the east’). The method of tuning it, extremely flexible in the 19th century, is now becoming stabilized. These modifications are due partly to the break-up of the Ottoman Empire, which has caused a rupture between Turkish and Arab cultures, and partly to the proliferation of teaching methods endeavouring to impose a single type of tuning, running from low to high: yaka = G; ‘ushayran = A; duka = d; nawa = g; kardan = c’. However, there are variants reintroducing tuning by 4ths. Thus what is described as ‘Aleppo tuning’ consists of: qarar busalik = E; ‘ushayran A; duka =d; nawa = g; kardan = c’. This latter structure is used in Turkey and Iraq. To answer the practical requirements of present-day notation, a treble clef followed by the figure 8 is used. This procedure has been much criticized by those in favour of using the bass clef. The tuning of the Turkish lute faithfully reflects the Arab type but in reverse, reading in descending order: gerdaniye = g’; neva = d’; dugah = a; asiran = e; kaba dugah = d (this last, more mobile pitch may equally settle upon G. This outdated tuning represents the ‘old school’ (eski akort), and has now been replaced by an ascending tuning – the ‘new school’ (yeni akort): A-B-e-a-d’-g’. Though it is now considered incorrect in the Syro-Egyptian area, and representative of the old Ottoman school, a tuning method in ascending order survives in Iraq. It consists of: yaka = d; ‘ushayran = e; duka = a; nawa = d’; kurdan = g’. The compass of the bichordal five-course ‘ud is just over two octaves; in Turkey, it is three octaves with the addition of a low course. Arabian instruments can achieve this by the addition of a sixth course.
(iv) Six-course ‘ud: Two kinds of six-course ‘ud exist: one has six pairs of strings, the other five pairs with an additional low string. The first was found by Jules Rouanet in North Africa towards the end of the last century; tuned inclusively it has since disappeared except in Libya, where it is still made but with different tuning. A similar instrument, found in Syria, is tuned C-E-A-d-g-c’. The instrument with five double strings and a single low one, however, is becoming increasingly usual from Istanbul to Baghdad. It has become common to place the additional string after the highest (or chanterelle). Its pitch is at the choice of the player; no rule is laid down. The presence of the extra string endows the instrument with a wider range and increased ease of playing, allowing the performer to run effortlessly through three octaves. The sixth course is also coming to be used as an intermittent drone, a new phenomenon.

(v) Seven-course ‘ud: Seven-course models, based on a complex system of tuning, were found in Egypt and Lebanon in the 19th century but have not been seen since 1900. There is one exception: the Tunisian, Fawzl Sayib, is a living master of the seven-course instrument in the six pairs and one low arrangement. A feature of this ‘ud was that it reversed the arrangement of strings, placing first the high and then the low strings on the neck from left to right. According to Mikha’il Mushaqa (1800-88), only four of the seven courses were played, the lowest course (jaharka) and the two highest (busalik and nihuft) being unused in performance.
5. Performance and Aesthetics

Turkish miniature, 18th cent. The strings of the contemporary ‘ud are twisted, or spirally reinforced. They are plucked with a plectrum (risha, ‘quill’) made of an eagle’s feather and held between thumb and index finger; a shell or plastic plectrum may be used instead. The technique calls for suppleness of the wrist as the plectrum strikes the strings in a simple fall, or combines risings and fallings. Certain teachers, such as Tawfiq al-Sabbagh, claim that a technique similar to the mandolin tremolo was once used. This may have disappeared, but another technique spread rapidly: the basm (‘imprint’), which was invented by the Egyptian Ahmad al-Laythi (1816-1913). It consists of substituting for the plectrum touches of the fingers of the left hand, plucking the strings, and introduces light and shade into the execution. Munir Bashir (Iraq) extended the technique by using the right hand too; he has made it one of the canons of present-day aesthetics of the ‘ud.

There are two schools or conceptions of performance. The first, or ‘Ottoman’, takes as its principle the ornamentation of the sound, produced by delicate glissandos of the fingers and slight vibratos. The touch of the plectrum on the string sets off a vibration which, in turn, gives rise to an effect of resonance, volume and controlled intensity. The plectrum does not interfere with the resulting sound. This produces an intimate style of playing, making the interiorized ‘ud a path to meditation. This approach was first promoted in Istanbul by Ali Rifat Cagatay (1867-1935) and Nevres Bey (1873-1937), then by Refik Tal’at Alpman (1894-1947) and Cinucen Tanrikorur (b 1938). It spread to Aleppo (Nash’at Bey, d c1930, and ‘Abd al-Rahman Jabaqji, b 1931), then was developed in Baghdad by Salman Shukur (b 1921), Jamil Bashir (1925-77) and Munir Bashir (b 1930). The last-named is the best-known contemporary ‘ud player; he has provided it with an instrumental repertory, the basis for recitals, eliminating its ancient association with song.

The second aesthetic approach is Egyptian. The volume is amplified by firm strokes of the plectrum, which makes the strings resonate; the result is a curiously dulled sound, akin to the nasal effect of Egyptian song. This calls for virtuosity in performance, which is conceived of as an exteriorizing factor. The finest proponents of this school have been Safar ‘Ali (1884-1962), Muhammad al-Qassabji (1898-1966) and Farid al-Atrash (1907-75), who, despite his melodramatic style, breathed a new vitality into the instrument. A synthesis of these two styles is taking place in Somalia, where the manner of performance combines extensive glissandos with the sonorous impact of the plectrum; the outstanding proponents of this style are Abdullahi Qarshe and ‘Umar Dhule.
6. Study of the ‘ud

With the appearance of new problems of theory, such as the 19th-century division of the octave into 24 quarter-tones, the ‘ud has entered a new phase. In the past it was not an ideal instrument for theoretical research, unlike the tanbur: ‘The ‘ud allows of theoretical demonstrations, but in an imperfect manner’ (Farabi). However, as the tanbur fell into disuse among Arabs during the 19th century, the ‘ud was substituted for theoretical reference. The present-day tendency towards a standardized teaching method based on a Western approach tries first to resolve the problems created by the use of microintervals not provided for in Western treatises, and second to produce teaching manuals adapted to the instrument’s evolution. The earliest such course to be published, in 1903, was by the Egyptian Muhammad Dhakir Bey (1836-1906): Tuhfat al-maw’ud bi ta’lim al-‘ud (‘The promise of the treasure, or the teaching of the ‘ud‘). Since then, various manuals have tried to ‘democratize’ the instrument, placing it within everyone’s reach and putting forward teaching rules that claim to be universal. They offer instruction in solfeggio and Western theory and give exercises on occidental or oriental modes. They all use Western notation, with modifications of key signature, and place before the student a large repertory, mostly of the 19th and 20th centuries. Notable among proponents of this method of teaching was Muhiddin Targan (1892-1967). This trend has been opposed, in the name of the elementary aesthetic rules of traditional Arab music (i.e. creative liberty and the development of the modal sense). But certain masters of the ‘ud owe something to these newer manuals. Two are outstanding for their instructional value, those by Fu’ad Mahfuz of Damascus (1960), which effects a synthesis between theoretical and practical, and ‘Abd al-Rahman al-Jabaqji of Aleppo (1982), the first accompanied by audio- and video-cassettes. In any event, it is still too early to analyse the consequences of written, standardized tuition in an instrument whose technique has been passed down individually and orally from master to pupil for over a thousand years.