HARP (čang, q.v.),a string instrument which flourished in Persia in many forms from its introduction, about 3000 B.C.E., until the 17th century. The original type was the arched harp as seen at Čoḡā Miš and on later third millennium seals (Figure 1a-c). Around 1900 B.C.E. they were replaced by angular harps with vertical (Figure 2) or horizontal (Figure 3) sound boxes. By the start of the Common Era, “robust, vertical, angular harps” (Figure 2), which had become predominant in the Hellenistic world, were cherished in the Sasanian court. In the last century of the Sasanian period, angular harps were redesigned to make them as light as possible (“light, vertical, angular harps,” Figure 4); while they became more elegant, they lost their structural rigidity. At the height of the Persian tradition of illustrated book production (1300 to 1600 C.E.), such light harps were still frequently depicted, although their use as musical instruments was reaching its end.

Extant harps from Ancient Egypt, Pazyryk (Central Asia, fourth century B.C.E.; Lawergren, 1990) and Japan (Figure 4b, eighth century C.E.) correspond closely with early representations of Persian harps, none of which actually has survived. By definition, a harp is a string instrument, the plane of the strings of which is perpendicular to the plane of the sound surface (for nomenclature, see Figure 2a). The largest part of the harp is a hollow box, the open side of which, facing the strings, is covered by the sound surface, which may be a stretched animal hide or a thin wooden panel. During the first millennia (B.C.E. and C.E.) the sound boxes of angular harps were either of even width or flared at the top (Figures 2a-c). The tops of the harps at Ṭāq-e Bostān (Figure 4; ca. 600 C.E.) curve forward, and many Central Asian examples from the following centuries show an increase in this tendency (e.g., Karomatov et al., 1987, Pls. 164, 167). The harps in Figures 2d-e represent this curved tradition, which is common in medieval book illustrations (Figure 7, bottom row).

Every harp has a narrow, wooden rib running along the center of its sound surface. One end of each string is tied to this rib while the other end is tied to a tuning device on the rod. For the first few millennia, the device used was a collar wound tightly around the rod, which was turned to adjust the strings (historical tuning methods are discussed in Lawergren, 1997b). During the Hellenistic period collars began to be replaced by rotatable pegs (similar to violin pegs), but the two tuning methods co-existed for a long time and both kinds are still shown in medieval book illustrations. While sinew strings were used for the Pazyryk harps (ca. 350 B.C.E.; Lawergren, 1990, p. 115), by the 14th century C.E. twisted silk, the same material that was used for lutes, had started to be used for Persian harp strings (Neubauer, 1990, p. 344).

One of the features that helps one to distinguish between arched and angular harps is the direction of the rod; while the rod of the former extrudes along the axis of the box (Lawergren, 1996, Figure 1), the latter’s rod is perpendicular to the axis. There is also a hole in the solid part of the box of the robust, angular harp for the rod to penetrate, the rod being non-circular to prevent it from rotating when the tuning collars are turned. The most obvious difference between arched and angular harps of the vertical kind is the number of strings they can accommodate; extant examples reveal that vertical, angular harps had between 15 and 25 strings (usually 21, a number that changed little between 1300 B.C.E. and 1600 C.E.), while arched harps usually had many fewer. While there are no extant examples of horizontal, angular harps (Figure 3a), Assyrian Palace reliefs (ca. 860-640 B.C.E.) may serve as a guide: they depict horizontal harps with about nine strings, and vertical harps, accurately, with 20-30 strings (Lawergren and Gurney, 1987, p. 51).

Most of the parts of the light, angular harp correspond to those on the robust model, but their dimensions and positions differ. For instance, the rod is thinner, and it is attached underneath the box rather than through a hole in it. This is most clearly exemplified by extant harps in Japan (Figure 4b; cf. reliefs in Persia, Figure 4a), which have a long tail descending under the box to support the rod. In order to improve rigidity a pin is inserted between the rod and the bottom of the box, thus fixing the rod to it at two points.


Third millennium B.C.E. This was the era of arched harps in Persia. It came to an end with the arrival of angular harps, ca. 1900 B.C.E. (Figure 5a). However, arched harps survived in India and diffused from there during the first millennium C.E. (Lawergren, 1995/6, pp. 244-45), including to Panjikent (Figure 1d).

Angular harps, one of the earliest depictions of which features in a terracotta plaque from Šahr-e Soḵta in eastern Persia (Tosi, 1968), were mostly vertical at first. Although its exact date of origin is unclear (since it was a surface find, the dating is uncertain and the excavator’s suggestion of 3000 to 2300 B.C.E. seems too early), together with the Čoḡā Miš seal (Figure 1a) it indicates that Persians may have then been the most active producers of harps in this era.

Second and first millennia B.C.E. The largest source of early depictions of angular harps is terracotta plaques made in Mesopotamia during the Isin/Larsa and Old Babylonian periods. The side view was especially popular (Figure 5a), but other views from the side and front were also depicted (Lawergren, 1997a, pp. 153-57, figs. 12-17). Terracotta plaques from Susa (1970-1650 B.C.E.) contain depictions of Persian angular harps that differ consistently from those in Mesopotamia: while their shapes are similar, the Persian harps are smaller in size.

Vertical harps are illustrated in Figure 5: Mesopotamian front and side views are shown in Figures 5a and b (see also Rashid, 1984, figs. 69, 70, 114, 115, front; figs. 62-70, side). In both of these depictions the Mesopotamian harp extends from the player’s navel to a point above his head by approximately the length of his head. There are no side view images of Persian vertical harps, but Figure 5c represents a front view (see also Spycket, 1992a, nos. 760 and 819), which reveals that the ElamitePersian harp is less than half as tall as the Mesopotamian harp. There is a similar discrepancy between the horizontal, angular harps from the two regions. The size of the Mesopotamian horizontal harp in the side view (Figure 3a) is approximately the same as the Mesopotamian vertical harp, but the Elamite-Persian instrument (Figure 3b; see also Spycket, 1992a, nos. 95-96) is less than half the size.

There are many Persian representations of harps dating from between the 12th and 7th century B.C.E. For example, large rock reliefs at Kul-e Fara, near Iḏa (Figure 6a), show ensembles of vertical and horizontal harps (Lawergren 1997a, fig. 26) which are as large as Mesopotamian ones. Equally large are the vertical and horizontal harps played outside the Persian city of Madaktu (Figure 6b), captured by Aššurbanipal ca. 645 B.C.E. Although harps were most often played in large homogeneous ensembles, the Arjan bowl depicts a heterogeneous group playing harps, lute, lyre, pipes, cymbals, and drums, as well as dancing, juggling and stilt-walking (Majidzadeh, 1992, p. 134).

Parthian and Sasanian periods. Parthian harps were typical of the Hellenistic genre, and so it is difficult to de-termine if they were derived from an indigenous tradition or from a western (Egyptian and Greek) one (Figure 2b). The Hellenistic harp had a straight box which sometimes expanded towards the top, and its rod had a wide girth.

In the Sasanian period the Hellenistic tradition continued, as demonstrated on a mosaic at Bišāpur (q.v.; fig. 2c). Harps on silver vessels (Figure 2d-e), which are dated slightly later, were as robust in form as Hellenistic harps, but their curved tops indicate Central Asian influence. Horizontal, angular harps are shown in Ṭāq-e Bostān (Fukai et al., 1972, Plates LI and LXV), and even as late as on an 8th-10th century C.E. silver plate (Figure 3c). The Ṭāq-e Bostān reliefs show a royal hunt accompanied by music; only harps were allowed on the king’s boat and on the accompanying barge, suggesting that they had become closely associated with royalty in the Sasanian period. As mentioned above, this period also witnessed the introduction of the light, angular harp, which became the most common type during the Islamic period.

The Islamic period. Light, angular harps remained popular in Persia well into the second millennium C.E., long after they had disappeared elsewhere. Harpists feature frequently in epic poetry which focuses on the earlier Sasanian period, and they are often depicted in their illustrations. The best-known story is perhaps that of Bahrām-e Gur and his mistress Āzāda, who played the harp while he hunted. The episode is related in Ferdowsi’s Šāh-nāma and Neẓāmi’s Haft Peykar (where the harpist is called Fetna; see tr. Meisami, 1989). Ettinghausen (1979, p. 29) and Fontana (1986, p. 9) have shown that the royal couple first appeared on Sasanian silverware and seals dating from 650-750 C.E., but the period in which they were depicted most frequently was between 1300 and 1600 C.E., when the tradition of book illustration flourished. In fact, these books provide the largest concentration of images of harps before the modern era, revealing a remarkable diversity of forms.

The workshops of western Persia, the main representatives of which were in Shiraz, Tabriz, Qazvin, and Isfahan (and occasionally in Baghdad), used a design which distinguished their harps from those produced by the workshops of eastern Persia. The first harps from western Persia depicted in illustrations (ca. 1300 C.E.; Simpson, 1979, figs. 51, 54, and 65) had pointed tops, short and straight tails, and plain, white sound boxes with the pin concealed under the player’s right hand and sleeve (Figure 7, top left). A century later the pin was visible with tassels often hung under the rod, and the box was covered with elaborate decoration (e.g., Figure 7, “Herat 1426”). In the next century, the products of western Iranian workshops adhered to much the same traits: the tops became slightly more elaborate; the tails remained relatively short to permit the harpist to sit; a pompom at the top replaced tassels under the rod; and the box was covered with decorative patterns (e.g., fig. 10, “Qazwin 1550”). Many harps were now depicted at an oblique angle, revealing the introduction of sets of holes on the sound surface. The 16th century marked the culmination of the western design, shortly after which it changed dramatically: for example, the harp from Isfahan in 1600 (Figure 7) lacks tail and pin, and features an innovatively shaped box.

Herat was the eastern center until the first decades of the 16th century, when Bukhara became more important. The eastern Persian harp flourished between about 1450 and 1530 C.E. Its distinctive design emerged in Herat and spread to Bukhara (fig. 10, “Herat 1524”). The side-view reveals a blunt top and a long sinuous tail that required the harpist to stand. The pin was integrated into the box, which was often decorated with six-pointed stars set in elaborate frames. Strings were frequently drawn in pairs, suggesting that it was a double-strung harp.

The Islamic period supplied the closing chapter in the world history of angular harps, which had been used in Persia itself since the beginning of the second millennium B.C.E. Čelebi’s report that ten čang makers and ten harpists lived in Istanbul in the middle of the 17th century is the last of such reports about their existence (Farmer, 1936, pp. 30-31). Why did angular harps become obsolete after an illustrious history extending more than four millennia? One answer is suggested in the little-known Aḥmadi’s satirical poem “A contest of string instruments,” written between 1400 and 1450 in Persia or Transoxania: in this poem, eight chordophones are personified, taking turns at boasting about their own virtues while berating the defects of others. The harp is criticized for easily falling out of tune and therefore requiring frequent adjustment (Bodrogligeti, 1987, p. 78, line 10; p. 79, line 3), a shortcoming of which none of the others is accused. Given its structural weakness, one would indeed expect problems with tuning. However, another possible reason for the demise of the harp was the rise of instruments belonging to the lute family; although they have only a few strings, each one can produce many pitches when shortened, thus covering the same range as a twenty-string harp. Bulky and fragile harps clearly could not compete against compact and sturdy lutes.

The question can also be turned around: why did angular harps last longer in the Islamic world than elsewhere? It is possible that book illustrations give an exaggerated image of their popularity, since they mostly illustrate tales of the fabled past, but medieval Persian texts, which probably give a more accurate portrait of actual harp usage, include references to harps in ways which suggest that contemporary poets and musicians actually used them. However, when musicians and their instruments are mentioned during the 15th and 16th centuries, the čang is rarely included (Thackston, priv. comm.). It is also worth observing that, of the eight instruments personified in Aḥmadi’s poem, the čang is the only one which boasts: “I am constantly in the company of kings” (Bodrogligeti, 1987, p. 78, line 7). It is likely therefore that angular harps had a prolonged life because they appealed to the Timurid and Safavid courts: the instrument served to link them to the Sasanian royal court, where it was valued highly.



Pierre Amiet, Age des échanges inter-iraniens, 3500-1700 av. J.-C., Paris, 1986.

Andras J. E. Bodrogligeti, “A Masterpiece of Central Asian Turkic Satire: Aḥmadi’s A Contest of String Instruments,” Ural-Altaic Yearbook 59, 1987, pp. 55-88.

Peter Calmeyer, “Zur Genese altiranischer Motive,” AMI, N.F. 6, 1973, pp. 135-52.

Malcolm A. R. Colledge, The Parthians, New York, 1967.

Pinhas Delougaz and Helene J. Kantor, Chogha Mish 1: The First Five Seasons of Excavations 1961-1971, Plates, Oriental Institute Publications 101, Chicago, 1996.

Éric De Waele, “Musicians and Musical Instruments on the Rock Reliefs in the Elamite Sanctuary of Kul-e Farah (Izeh),” Iran 27, 1989, pp. 29-38.

Richard Ettinghausen, “Bahram Gur’s Hunting Feats or the Problem of Identification,” Iran 17, 1979, pp. 25-31.

Henry G. Farmer, “Turkish Instruments of Music in the Seventeenth Century,” JRAS, January, 1936, pp. 1-43.

Idem, “The Music of Islam,” in The New Oxford History of Music: Ancient and Oriental Music, ed., Egon Wellesz, London, 1957, I, pp. 421-77.

Idem, Islam:Musik-geschichte in Bildern, III.2, Leipzig, 1966.

Maria V. Fontana, La leggenda di Bahrām Gur e Āzāda: Materiale per la storia di una tipologia figurative dalle origini al XIV secolo, Naples, 1986.

Shinji Fukai, J. Sugiyama and Kiyoharu Horiuchi, Taq-i Bustan 1, The Tokyo University Iraq-Iran Archaeological Expedition 13, Tokyo, 1972.

Ann C. Gunter and Paul Jett, Ancient Iranian Metalwork in the Arthur M. Sackler Gallery and the Freer Gallery of Art, Washington, 1992.

Kenzo Hayashi, Shigeo Kishibe, Ryoichi Taki, and Sukehiro Shiba, Musical Instruments in the Shôsôin, Tokyo, 1967.

Faizulla M. Karomatov, V. A. Meskeris and Tamara S. Vyzgo, Mittelasien: Musikgeschichte in Bildern, II.9, Leipzig, 1987.

Bo Lawergren, “The Ancient Harp from Pazyryk,” Beiträge zur allgemeinen und vergleichenden Archäologie 9/10, 1990, pp. 111-18.

Idem, “The Spread of Harps between the Near and Far East during the First Millennium A.D.,” Silk Road Art and Archaeology 4, 1995/96, pp. 233-75.

Idem, “Harfe,” Die Musik in Geschichte und Gegenwart 4, ed. L. Finscher, Kassel/Stuttgart, 1996, pp. 39-62.

Idem, “Mesopotamien,” Die Musik in Geschichte und Gegenwart 6, ed. L. Finscher, Kassel/Stuttgart, 1997a, pp. 143-171.

Idem, “To Tune a String: Dichotomies and Diffusions between the Near and Far East,” in B. Magnusson, S. Renzetti, P. Vian, and S. J. Voiçu, eds., Vltra Terminvm Vagari: Studi in onore di Carl Nylander, Rome, 1997b, pp. 175-92.

Y. Majidzadeh, “The Arjan Bowl,” Iran 30, 1997, pp. 131-44.

Julie S. Meisami, “Fitnah or Azadah? Nizami’s Ethical Poetic,” Edebiyat, N.S. 1/2, 1989, pp. 41-75.

Eckhard Neubauer, “Der Bau der Laute und ihre Besaitung nach arabischen, persischen und türkischen Quellen des 9. bis 15. Jahrhunderts,” Zeitschrift für Geschichte der arabisch-islamischen Wissenschaften 8, 1990, pp. 279-378.

W. Orthmann, Der alte Orient, Propyläen Kunstgeschichte 14, Berlin, 1975.

Christian Poche, “Wann,” The New Grove Dictionary of Musical Instruments, 3 vols., London, 1984, p. 838.

Edith Porada, The Art of Ancient Iran: Pre-Islamic Cultures, New York, 1965.

Idem, “Discussion of a Cylinder Seal, Probably from Southeast Iran,” Iranica Antiqua 23, 1988, pp. 139-143.

Subhi A. Rashid, Mesopotamien: Musikgeschichte in Bildern, II.2, Leipzig, 1984.

Basil W. Robinson, “The Turkman School to 1503,” in B. Gray, ed., The Arts of the Book in Central Asia 14th-16th Centuries, Paris, 1979, pp. 215-47.

George D. Sawa, Music Performance Practice in the Early ʿAbbāsid Era 132-320/750-932, Toronto, 1989.

Marianna S. Simpson, The Illustrations of an Epic: The Earliest Shahnama Manuscripts, New York, 1979.

Agnès Spycket, Les figurines de Suse 1, Les figurines humaines: Mémoires de la Délégation archéologique en Iran 52, Paris, 1992a.

Ibid, “Popular Art at Susa: Terracotta Figurines,” in P. O. Harper, J. Aruz, and F. Tallon, eds., The City of Susa: Ancient Near Eastern Treasures in The Louvre, NewYork, 1992b, pp. 183-96.

Mario Tosi, “Excavations at Shahr-i Sokhta, a Chalcolithic Settlement in the Iranian Sistan: Preliminary report of the First Campaign, October-December 1967,” East and West 18, 1968, pp. 9-16.

(Bo Lawergren)

Originally Published: December 15, 2003

Last Updated: March 6, 2012

This article is available in print.
Vol. XII, Fasc. 1, pp. 7-13