ḠAZNĪ (or Ḡazna, Ḡaznīn), province and city in southeastern Afghanistan, the latter situated 136 km south of Kabul at an altitude of about 2,200 meters (lat 33°34´ N, long 68°27´ E).

i. Historical geography.

ii. Monuments and inscriptions.



The town of Ḡaznī dominates the route which, from Kabul to Qandahār (Kandahar), skirts southward around the central highlands of Afghanistan and over easy passes, either through the Gardīz and Ḵōst basins, and over the Batay pass and the Točī valley, or further south through the Gomal valley, which provides a relatively easy access to the middle course of the Indus. At the southernmost end of a jagged plateau joining the highlands, the site of the present city is reached over a steep hill dominating the Ḡaznī river, which flows from north to south towards the Kaṭawāz (Āb-e Īstāda; q.v.) basin. Here a small market-town was established a long time ago, perhaps already mentioned by Ptolemy (VI, 18, 4: Gázaca or Gāzaca), and at any rate clearly identifiable with the Ho-si-na where the Chinese pilgrim Hsüan Tsang stopped in 629 (Beal, II, p. 283). In the 10th century, Ḡaznī was described in Arabic geographical accounts as a busy commercial center securing Central Asian and Iranian relations with India (Le Strange, Lands, p. 348). Eṣṭaḵrī (q.v.), who described it as having a river but no gardens, wrote that no other city in the region was richer in commerce and merchandise, for it was the “port” of India (p. 280). Ebn Ḥawqal (q.v.), whose Ketāb ṣūrat al-arż was completed in 988, but who probably visited the region in 969, also knew it as a trading center with India, frequented by numerous Hindu merchants. But he describes it as having greatly suffered since its siege and capture in 966 (in fact 961) by Alptigin (q.v.), the first Turkish governor who became almost autonomous in the region (Bosworth, Ghaznavids, pp. 37-38). It was, however, from this period on that the city greatly expanded, ruled by a Turkish soldiery from which there soon emerged (with Sebüktigin in 977) the Ghaznavid dynasty. The city’s commercial function was enhanced by its role as a capital, as the headquarters of regularly organized expeditions sent on the conquest of India, and as a major cultural center. With its by now considerable population, the city had largely expanded towards the hills and the plateau, on both sides of the river, and also into the valley and the plain, as attested by the considerable ravages caused by a flood in 422/1031 in the markets and caravansaries (Bosworth, Ghaznavids, p. 139, quoting Bayhaqī). The water supply for the city and the irrigable lands necessary for its provisions was secured by constructing large dams which have survived and been renovated recently (see BAND; Balland, p. 241).

The city’s prosperity came to an abrupt and brutal end when it was sacked and set on fire in 1151 by the Ghurid sultan ʿAlāʾ-al-Dīn Ḥosayn Jahānsūz (q. v.). The return of the last Ghaznavids brought a degree of renewal for a few decades, and Yāqūt, who reported on events on the eve of the Mongol invasion, still considered it a great city (Yāqūt, tr., p. 406). However, the Mongols dealt the final blow and destroyed it again in 1221. In 1332 or 1333, Ebn Baṭṭūṭa reported that “most of it is devastated.” The inhabitants spent the winter in Qandahār because of the cold. From then on the bulk of the population moved south of the old capital, to a “small town on a river flowing under the citadel,” where the traveler stayed (Ebn Baṭṭūṭa, III, pp. 88-89). This must already have been the present “old town,” on a hill about a hundred meters above the left bank of the river where it enters the plain. From then on the descriptions remain unchanged for several centuries. Ḡaznī became merely a modest local center. Bābor (q.v.), in 1504-5, considered it as “a very humble place,” merely surviving because of the river which fed four or five mills and as many farming villages. He found it strange that the masters of Hindustan and Khorasan could have made it their capital (Bābor-nāma, tr. Beveridge, pp. 218-19). This was also the impression conveyed by the first European travelers in the 19th century. In 1832, according to Charles Masson, it consisted of no more than about a thousand houses, and its bāzār was mediocre and poorly stocked (Masson, II, p. 219). This is confirmed by Godfrey Thomas Vigne (p. 127) who, in 1836, also mentions a “modest” bāzār (see also Mohan Lal, pp. 329-30, who passed through in 1833, and Bellew, 1862, p. 185 ff.). One of the main functions of the city, perched on top of its hill and surrounded by a wall, was to house a relatively large garrison of about two thousand men (Vigne, pp. 122-23, 127), an essential element for controlling the road and the regional tribes. However, since the town had moved to the edge of the plateau after abandoning the Ghaznavid city, it was dominated by the surrounding hills and therefore unable to resist a proper siege (Masson, p. 221).

A new phase in the city’s development started in the 1920s, with the beginnings of the country’s modernization and the growth of road transport. Cars began to appear on the Kabul-Qandahār route, which runs along the foot of the village, on the other bank of the river. Ḡaznī then became a stopping-off point. A suburb rapidly appeared along the road, with a new bāzār for supplying travelers and vehicles; but it was still, until 1958, insignificant in comparison with the old city. At that date, the Ministry of Public Works started building a new city in the plain south of the old one on plots of 800 to 1,000 sq. m. This included an administrative center and public buildings, with new bāzārs on both sides of the river and a large Friday mosque. It rapidly monopolized all the central functions, which now deserted the old city. From the early 1970s on, the shops in the bāzārs of the old city became dilapidated or closed down, and crossroads (čahār sūqs) turned into empty spaces. Only the military quarter and the citadel remained unchanged. However, poor immigrants began to settle in the old city quarters, and by 1972, this area contained four-fifths of the town’s homes, with a density of some 700 inhabitants per hectare, as against 60 to 80 inhabitants per hectare in the new city. Since 1970, however, the latter has also become more densely populated, and the newly built plots have been reduced in size to 300-400 sq. m., leading to a density of about 200 homes per hectare. Apart from the affluent classes, who have all left the old city, these new quarters house newcomers (artisans and tradesmen) coming from nearby Kaṭawāz. They own a considerable part of the 2,400 or so shops, and regularly cater to the vehicles stopping on their way, as well as to the neighboring countryside. The total population of the town was in the order of 20,000. This type of urban evolution, marked by the persistence of an old city preserving a large part of its population next to a new city which has captured all its central functions, is represented by numerous towns in Afghan Turkestan (Grötzbach, 1979, pp. 26-27), but Ḡaznī is the only example in southern Afghanistan south of the Hindu Kush. This exception is evidently due to the particular morphology of the old city which, being situated on a steep hill, did not lend itself to a transformation in situ.



D. Balland, “Passé et présent d’une politique des barrages dans la région de Ghazni,” Stud. Ir. 5, 1976, pp. 239-53.

S. Beal, Si-Yu-Ki: Buddhist Records of the Western World, Translated from the Chinese of Hiuen Tsiang (A. D. 629), 2 vols. in 1, London, 1884.

Gazetteer of Afghanistan VI, pp. 183-201.

E. Grötzbach, “Zur jungen Entwicklung afghanischer Provinzstädte: Ghazni und Mazar-i Sharif als Beispiele,” Geographische Rundschau 27, pp. 416-24.

Idem, Städte und Basare in Afghanistan: Eine stadtgeographische Untersuchung, Beihefte zum Tübinger Atlas des Vorderen Orients B 16, Wiesbaden, 1979.

C. Masson, Narrative of Various Journeys in Balochistan, Afghanistan, and the Panjab, 3 vols., London, 1842; repr., Karachi, 1974.

E. Merçil, Gazneliler devleti tarihi, Ankara, 1989.

Mohan Lal, Travels in the Panjab, Afghanistan and Turkistaņ, London, 1846; repr. Patiala, 1971.

G. T. Vigne, A Personal Narrative of a Visit to Ghuzni, Kabul, and Afghanistaņ, 2nd ed., London, 1843, repr. Lahore, 1982.

Yāqūt b. ʿAbd-Allāh Ḥamawī, Moʿjam al-boldān, tr. C. Barbier de Meynard as Dictionnaire géographique, historique et littéraire de la Perse…, Paris, 1861.




The earliest known monuments of Ḡaznī belong to the Ghaznavid period (366-583/977-1187), the best representative of which are the two minarets standing east of the citadel, close to two large mounds resembling mosques (Bombaci, 1959, p. 7; Figure 1). Each minaret bears a dedicatory inscription containing respectively the names of Masʿūd III (492-509/1099-115) and Bahrāmšāh (512-52/1118-157). They are built of brick and present two similar polygonal star-shaped basements that were originally surmounted by cylindrical shafts (Vigne, p. 125; Survey of Persian Art III, fig. 337; PLATE I). The shafts collapsed between the end of the 19th and the beginning of the 20th century, and the remains were subsequently covered by a tin roof.

Three different types of inscriptions are carved on Masʿūd’s minaret, whose decoration remained incomplete when he died. The first inscription, in bordered kufic, decorates the upper part of the basement. The name and titles of the sultan follow the besmellāh formula (q.v.). Next is a fragmentary cursive inscription on a narrow band framing some panels and containing the entire chapter 48 (Sūrat al-fatḥ) of the Koran. The last one, in square kufic, containing the name of the sultan and some of his titles, decorates the lower part of the basement (Pinder-Wilson, 1985, pp. 90-91).

The only inscription carved on the upper part of Bahrāmšāh’s minaret is in bordered kufic similar to that of Masʿūd’s minaret; it starts with the basmala followed by the sultan’s name and titles (Sourdel-Thomine, 1953).

During the excavation campaigns (1957-64) the Italian Archaeological Mission brought to light the remains of a magnificent palace in the plain of Dašt-e Manāra at some 300 m east of Masʿūd’s minaret (Scerrato). The archaeologists ascribed the palace to Masʿūd III both on the basis of its proximity to his minaret and the two cursive inscriptions carved respectively on a fragmentary meḥrāb arch and a marble architectural element. The first contains the name of the sultan, the latter that of the architect and the date of construction (Ramażān 505/March 1112; Bombaci, 1966, pp. 3-4), i.e., towards the end of Masʿūd III’s reign. The palace, with a four-ayvān courtyard, built of brick with marble panels and pavements, constituted a real royal town with a mosque, some residential, commercial, and military quarters, public areas, and gardens. Of the more than 510 original marble panels (70 cm high) of the base of the walls surrounding the main court, 44 came to light. Each panel has three parts; the upper part, about 10 cm high, bears an inscription in Persian verses carved in foliated kufic (one of the oldest examples of Persian epigraphy; see epigraphy iii). The central part, about 58 cm high, shows elaborate palmettes that fill the typical Ghaznavid trefoil arches. The lower part, approximately 6 cm high, bears a narrow band of vegetal scrolls. The ornament of these marble panels sometimes shows quadrupeds and birds (Bombaci, 1966).

Literary sources describe the intense building activity of the Ghaznavid rulers, but no other remains have been discovered until now (Bombaci, 1958). Several factors contributed to the rapid disappearance of the buildings: numerous devastating assaults (beginning with the fire set by the Ghurid ʿAlāʾ-al-Dīn Ḥosayn in 545/1150-51), powerful earthquakes (esp. in 1505, 1832, 1842, 1874, and 1902; Stenz, pp. 41-50), the tendency of each Ghaznavid ruler to abandon the residence of his predecessor, and the effect of cold climate on brick-structures. The historian Abu’l-Fażl Bayhaqī (q.v.) reports that the residence of Masʿūd I was in ruin only twenty years after its construction (ed. Fayyāż, p. 499). The traveler Ebn Baṭṭūṭa visited Ḡazna in the 14th century and was surprised by the desolate landscape around it (III, p. 88). In 910/1504 the Mughal Ẓahīr-al-Dīn Moḥammad Bābor seized Ḡaznī and described it as a “very humble place.” He mentioned the tombs of Maḥmūd, Masʿūd, Ebrāhīm, as well as many anonymous “blessed” ones (Bābor-nāma, tr., Beveridge, pp. 218-19).

From 1957 to 1966 the Italian Archaeological Mission uncovered many tombs scattered over the several small cemeteries and shrines (zīārāts) located in particular to the east of the citadel. The white marble monuments distinguish a conspicuous group of individual tombs, thus constituting evidence for the uninterrupted evolution in the construction of tombs and the related inscriptions since the beginning of the Ghaznavid period (Giunta). The tombs consist of two or more superimposed and scaled-down architectural elements forming a truncated pyramid. Each tomb is composed of a basement and a crested top element connected by one or more intermediate element.

The tombs of the Ghaznavid period, generally without cenotaphs, consist of low basements surmounted by prismatic elements and a rich variety of stepped socles and crested top components. In contrast, the funerary architecture of the Ghurid (q.v.) period was always provided with high cenotaphs, while the prismatic elements changed in height, width, and length, and the stepped socles disappeared. Since the 15th century, the tombs became simpler and the intermediate elements progressively disappeared. The funerary architecture of Ḡaznī has no close parallels in any other region of the Islamic world. It is noteworthy, however, that pictures of similar tombs can be found in some 13th-century manuscripts (Rice, pls. iv-vii; Ateş, pls. 11, 15, figs. 31, 42-44; Melikian-Chirvani, figs. 62-63, 65).

The inscriptions, carved in relief, generally occupy the four sides of each element of the tomb. The epitaphs follow a standard scheme which changes according to the dynasties. During the Ghaznavid period the most complete formula includes the besmala, the šahāda, the formula hāḏā qabr… (this is the tomb of…), the name of the deceased together with his genealogy and titles, the date of death, some koranic verses, and invocations for the deceased. The tombs of the Ghurid period lack the name of the deceased, while the šahāda occupies the two longitudinal faces of the top element. The āyat al-korsī (Koran 2:255) was often repeated on the elements of the tomb. In late tombs the epitaphs generally contain the name of the deceased and some koranic verses. The majority of Ḡaznī tombs are dateless, but they can be dated on the basis of those bearing dates or clear attributions.

The tombs are of great interest both for their structure and their ornamental and epigraphic features. Under the Ghaznavids kufic reached very eleborated forms acquiring a high decorative character. Four types of kufic were adopted: simple kufic with some decorative devices (since the end of the 10th cent.); square kufic (since the 11th cent.), bordered kufic in which the letters are often prolongated and generate two half palmettes or half leaves (since the mid-11th cent.); and foliated kufic whose angular letters bear leaves, scrolls, and half palmettes (since the beginning of the 12th cent.). Many ornamental elements are freely inserted in the spaces above or beside the letters. Kufic is not documented in the funerary inscriptions of the Ghurid period.

Since the first half of the 11th century the Ghaznavids also used the cursive script for funerary epigraphy as in one of the twenty-four inscriptions carved on the tomb of Maḥmūd (d. 421/1030) and in that of an unknown person who died in 447/1055. Both inscriptions contain six lines framed by a trilobed arch (PLATE II). These two inscriptions, together with another one carved on marble from the Ghaznavid mosque at Rāja Gīrā in the Swāt Valley and dated 440/1048-49 (Nazir Khan, pp. 153-66), are probably the earliest monumental cursive inscriptions in the Islamic world.



F. R. Allchin and N. Hammond, eds., The Archaeology of Afghanistan from Earliest Times to the Timurid Period, London, New York, and San Francisco, 1978.

O. Aslanapa, Turkish Art and Architecture, London, 1971.

A. Ateş, “Un vieux poème romanesque persan: récit de Warqah et Gulshāh,” Ars Orientalis 4, 1961, pp. 143-52, pls. 1-15.

A. D. H. Bivar, “Naghar and Iryāb: Two Little-Known Islamic Sites of the North-West Frontier of Afghanistan and Pakistan,” Iran 26, 1986, pp. 131-38, fig. 1, pl. i.

S. S. Blair, The Monumental Inscriptions from Early Islamic Iran and Transoxiana, Leiden etc., 1992.

A. Bombaci, "Ghaznavidi,” in Enciclopedia Universale dell’Arte VI, 1958, pp. 5-16.

Idem, “Introduction to the Excavations at Ghazni: Summary Report on the Italian Archaeological Mission in Afghanistan,” East and West 10, 1959, pp. 3-22, figs. 1-17.

Idem, “Les Turcs et l’art ghaznavide,” First International Congress of Turkish Arts, Ankara 1959, Ankara, 1961, pp. 67-70.

Idem, The Kūfic Inscription in Persian Verses in the Court of the Royal Palace of Masʿūd III at Ghazni, Rome, 1966.

E. Diez, Persien: Islamische Baukunst in Churāsān, Gotha, 1923.

S. Flury, "Das schriftband an der Türe des Maḥmūd von Ghazna (998-1030),” Der Islam 8, 1918, pp. 214-27.

Idem, “Le décor épigraphique des monuments de Ghazna,” Syria 6, 1925, pp. 61-90, figs. 1-11, pls. v-xxiv.

E. Galdieri, A Few Conservation Problems Concerning Several Islamic Monuments in Ghazni (Afghanistan): Technical Report and Notes on a Plan of Action, Rome, 1978.

R. Giunta, Les inscriptions de la ville de Ġaznī (Afghanistan), Ph.D. diss, Université de Provence, 1999.

J. A. Godard, “L’inscription du minaret de Masʿūd III à Ghazna,” Athār-è Īrān 2/1, 1936, pp. 367-69, fig. 236.

A. Grohmann, “The Origin and Early Development of Floriated Kūfic,” Ars Orientalis 2, 1957, pp. 183-213, figs. 1-34, pls. i-x.

A. Grohmann, Arabische Palaographie II: Das schriftwesen. Die lapidarschrift, Vienna, 1971.

ʿA.-Ḥ. Ḥabībī, Tārīḵ-e ḵaṭṭ wa neveštahā-ye kohan-e Afḡānestān, Kabul, 1971.

G. J. Jalālī, Ḡazna wa ḡaznavīān, Kabul, 1351/1932.

A. S. Melikian-Chirvani, “Le roman de Varqe et Golšāh,” Arts Asiatiques 22, 1970, pp. 1-262, figs. 1-65.

D. Monchi-Zadeh, "Notes on the Kūfic Inscription of the Royal Palace of Masʿūd III at Ghazni,” Orientalia Suecana, 16, 1967, pp. 113-26.

A. A. Naimi, “Les monuments historiques et les mausolées de Ghazni,” Afghanistan 7/2, 1952, pp. 9-18.

M. Nazir Khan, “A Ghaznavid Historical Inscription from Uḍegrām, Swāt,” East and West 35, 1985, pp. 153-66, figs. 1-7.

R. Pinder-Wilson, Ghaznevid and Ghurid Minarets, Kabul, 1981.

Idem, “The Minaret of Masʿūd III at Ghazni,” Studies in Islamic Art, London, 1985, pp, 89-102, pls. 1-7.

G. A. Pugacenkova, "La genèse centre-asiatique des minarets indiens,” Cahiers d’Asie Centrale 1-2: Inde-Asie centrale, Routes du commerce et des idées, Aix en Provence, 1996, pp. 127-32.

D. S. Rice, "The Oldest Illustrated Arabic Manuscript,” BSO(A)S 22, 1959, pp. 207-20, pls. i-ix.

M. Reżā, Rīāż al-alwāḥ (Inscriptions from Ghazni), Kabul, 1967.

U. Scerrato, “The First Two Excavation Campaigns at Ghazni, 1957-1958.

Summary Report on the Italian Archaeological Mission in Afghanistan,” East and West 10, 1958, pp. 23-55, figs. 1-60.

J. Sourdel-Thomine, "Deux minarets d’époque seljoukide en Afghanistan,” Syria 30, 1953, pp. 108-36, figs. 1-4.

Idem, “L’art ġūride d’Afghanistan à propos d’un livre récent,” Arabica 7, 1960, pp. 273-80.

Idem, “Ghaznawides: art et monuments,” EI2 II, pp. 1053-55.

Idem, “Ḳabr,” ibid., IV, pp. 352-55. Idem, “A propos du cénotaphe de Mahmūd à Ghazna (Afghanistan),” in A. Daneshvari, ed., Essays in Islamic Art and Architecture in Honour of Katharina Otto-Dorn, Malibu, Calif., 1981, pp. 127-35, figs. 1-4.

E. Stenz, “Earthquakes in Afghanistan,” Afghanistan 1, 1946, pp. 41-50.

Idem, "A Manuscript from the Library of the Ghaznavid Amir ʿAbd al-Rashid,” in R. Pinder-Wilson, ed., Paintings from Islamic Lands, London, 1969, pp. 17-21.

Y. Tabbaa, "The Transformation of Arabic Writing: Part 2. The Public Text,” Ars Orientalis 26, 1994, pp. 117-47, figs. 1-30.

G. T. Vigne, A Personal Narrative of a Visit to Ghazni, Kabul, and Afghanistan and of a Residence at the Court of Dost Mohamed with Notices of Runjit Sing, Khiva and the Russian Expedition, Lahore, 1982.


(Xavier de Planhol, Roberta Giunta)

Originally Published: December 15, 2000

Last Updated: February 3, 2012

This article is available in print.
Vol. X, Fasc. 4, pp. 384-388