ARCHITECTURE vi. Safavid to Qajar Periods



vi. Safavid to Qajar Periods

Iranian architecture from the 16th to the 19th centuries is, not surprisingly, dominated by the Safavids. Though no accurate checklist has been drawn up, it is clear that within the present political borders of Iran several hundred buildings datable between 907/1502 and 1138/1725 survive. No previous dynasty can rival this total. Some conceptual framework is therefore needed to encompass such a varied mass of material, and this framework will take precedence in the following account over detailed analyses of individual buildings. Three broad headings suggest themselves: the political context of Safavid architecture, the type, quantity, distribution, and time scale of this architecture, and its stylistic development.

Political context. At first sight these times would seem auspicious for great architectural projects: many shahs reigned for a period of between 25 and about 50 years. But despite this continuity, sustained architectural campaigns occurred only sporadically, for several of these monarchs (e.g., Esmāʿīl I or Ṭahmāsp) failed to provide the necessary impetus from the top. As a result the period of 1500 1590 is remarkably devoid of great architecture. With the accession of Shah ʿAbbās this situation changed dramatically, and the reason was precisely that not only the monarch but his courtiers too—physicians (e.g., Masjed-e Ḥakīm, Isfahan), generals, amirs, chamberlains, major domos and governors—began to build. Their joint activity and emulation sufficed to transform Isfahan beyond recognition. Beyond the orbit of the court there may have existed a class of merchant patrons whose financial support would help to explain the large number of lavish Safavid bazaars (Qazvīn, Kermān, Kāšān and Qom apart from Isfahan itself), but the shah himself still played the pre eminent role, and his foundations were inspired by motives as much political and economic as religious. The establishment of Shiʿism as the national creed encouraged increased veneration of the tombs of saints, but Ardabīl enjoyed unique status as a dynastic necropolis (see M. E. Weaver, Preliminary Study of the Conservation Problems of Five Iranian Monuments, UNESCO, Paris, 1970, suppl., Paris, 1971) and the Mašhad shrine (see ĀSTĀN-E QODS) was lavishly embellished, apparently to attract pilgrims (and their wealth) there rather than to Mecca, then in Ottoman hands. Many Sunni monuments in Khorasan (e.g., Torbat-e Jām shrine) profited from the patronage of Shah ʿAbbās, who was perhaps trying thereby to win over potential opponents. Economic motives account for the extensive network of trade communications which he built up throughout Iran. Urban caravanserais complemented the rural ones and a major port was developed at Bandar-e ʿAbbāsī.

The imperial aspirations of ʿAbbās and his eagerness to establish contacts with Europe explain much about his architecture. From 1005/1596 he sought to make Isfahan rival Istanbul and Delhi, which at that very time were undergoing an ostentatious face lift that clearly proclaimed imperial aspirations. Deprived of the particular natural advantages of Istanbul, with its built in vistas, he chose instead to expand his city, taking over a vast new acreage of virgin soil and creating entirely man made prospects within this area. This result was urban development on a modern scale (see the various articles on Safavid Isfahan in Journal of Iranian Studies 7/3 4, 1974). This expresses the political fact that in the Safavid period the capital gradually became central to government and society. Fine religious or palatial architecture was rarely produced in the provinces unless the monarch himself or the court was responsible for it.

Type, quantity, distribution. The massive output of Safavid monuments gives us—as for no other period—a truly representative selection of buildings. Generalizations can be tested in depth. Stylistic developments can be pinpointed, and even a respectable oeuvre assembled for individual artists (e.g., Moḥammad Reżā Emāmī; see A. Godard “Muḥammad Riḍā al Imāmī,” Atār-e Īrān 3/2, 1938, pp. 267 74; D. Pickett, “Inscriptions by Muḥammad Riḍā al Imāmī,” Iran 22, 1984, pp. 91 102). Major blanks in the chronology (e.g., between 1520 and 1590) can be identified as gaps in production and not confused with the vagaries of survival. Areas of active and of sluggish production emerge, as does the order of popularity of the various building types. It quickly becomes clear that this architecture has its roots in everyday life. Numerous Safavid buildings are situated unpretentiously in villages or small towns, and even those in the major cities are most often located in quarters that are essentially village or small town units transported en bloc into an urban setting. This is especially true of Isfahan (best short survey: A. Godard, op. cit., 2/1, 1937, pp. 7 176).

The areas where Safavid buildings cluster most densely differ significantly from the centers of earlier medieval architecture. Khorasan and Azerbaijan are neglected and even the earlier Safavid capitals of Tabrīz and Qazvīn saw little major construction. In Iran proper only Isfahan, and to a lesser extent Mašhad and Kermān, offer the chance to study numerous examples of ambitious Safavid work. For the rest of the country Safavid is more a convenient dynastic label than a precise descriptive term connoting a distinctive style. Safavid architecture had a certain cachet in neighboring lands, as shown by Iraq with its Shiʿite shrines, the Caucasian provinces and Uzbek Bukhara. Further east the tiled buildings of Qandahār, Lahore, and Mūltān seem to reflect contemporary Iranian inspiration indirectly.

Although the Safavid period did not generate brand new types of buildings, its priorities did not coincide with those of previous centuries. Seventeenth century Isfahan is a splendid proof of how thoroughly Shah ʿAbbās understood the psychological dimension of lavish public architecture, but large new mosques are a rarity in this period, in marked contrast to Ottoman and Mughal practice. Instead, the characteristic Safavid activity in religious—as distinct from secular— architecture is repair work. This trend is already evident in the best buildings associated with Esmāʿīl I (r. 907 30/1501 24); the Hārūn-e Welāyat Mausoleum (918/1512 13) and the Masjed-e ʿAlī (929/1522 23) (L. Honarfar, Ganjīna ye āṯār-e tārīkī-e Eṣfahān, Isfahan, 1344 Š./1965). Even in capitals like Tabrīz and Qazvīn, and at first Isfahan, it was apparently considered sufficient to refurbish the existing jāmeʿ. Often a new portal, or a facade screening the earlier building, claimed the whole structure for the Safavid patron (Masjed-e Jomʿa in Isfahan). Few important mosques or shrines in the country lack traces of Safavid work.

It is difficult to pinpoint the trends underlying these repairs. The attitude to the past is a case in point. Safavid inscriptions on the pre Islamic monuments (e.g., Persepolis and Bīsotūn) perhaps presage that wholesale adoption of and identification with ancient Iran that later characterized the Qajars, but there are not enough inscriptions to clinch the point. Repair work certainly familiarized Safavid craftsmen with earlier styles and possibly molded their own style on occasion (e.g., Safavid additions at Mahān, see M. E. Bāstānī Pārīzī, Rāhnamā ye āṯār-e tārīḵī-e Kermān, Tehran, 1335 Š./1965) or Kūhpāya (M. Siroux in Annales Islamologiques 6, 1966, pp. 153 54). But extreme conservatism rather than conscious antiquarianism may explain such imitations. The available evidence indicates that it was precisely the shahs who avoided large building projects—Esmāʿīl I (907 30/1501 14), Ṭahmāsp I, (930 84/1524 76), Ṣafī (1038 52/1629 42), Solaymān (1072 1105/1666 94), and Sultan Ḥosayn (1105 35/1694 1722)—whose reigns also witnessed a disproportionate amount of repair work. The last two names suggest that the gradual enfeeblement of the dynasty increasingly forced architectural patronage to confine itself to relatively trivial projects.

Among the various building types of cultic architecture small mosques were a standard feature, and in Isfahan, where most are to be found, were built mostly at the behest of courtiers (Honarfar, op. cit., passim). Hypostyle, domed square and ayvān mosques are all common. Large, purpose built madrasa structures—as distinct from composite foundations, now not always recognizable as such, or mosques which served at least in part as madrasas—continued to be popular but were now supplemented by a smaller and more intimate type of building, something perhaps more akin to the original concept of the madrasa. The courtyard remained, but was so reduced in size that the building reverted to an essentially domestic scale. Most Safavid religious buildings, however, are shrines. These greatly outnumber secular funerary buildings; the shahs themselves showed the way by being buried in shrines at Ardabīl, Qom, and Kāšān. Existing shrines at Mašhad (see Āstān-e Qods; Survey of Persian Art, chap. 39L), Qom and elsewhere were greatly enlarged and the adoption of Shiʿism by the state encouraged the proliferation of emāmzādas. These buildings had spacious interiors so that they could readily function as places of worship and pilgrimage (e.g., Ḵᵛāja Rabīʿ near Mašhad) and the Qadamgāh near Nīšāpūr (for bath buildings, see E. Diez, Churasanische Baudenkmäler, Berlin, 1918).

An unexpected burst of activity in secular architecture marks the 17th century. Bridges which have wider functions than carrying traffic were built, reviving Sasanian custom (Survey of Persian Art, chap. 39M). There are large urban bazaars (for that of Isfahan cf. A. A. Bakhtiar in Proceedings of VIth International Congress of Iranian Art and Archaeology, Tehran, 1976). Scores of urban and rural caravanserais (M. Siroux, Caravanserails de l'Iran, Cairo, 1949, passim) and thousands of pigeon towers (in the Isfahan, Golpāyagān, and Kermān areas) were erected (cf. E. Beazley, “The Pigeon Towers in Iṣfahān,” Iran 4, 1966, pp. 105-9). Town planning develops on a scale hitherto unequalled in Iran (E. E. Beaudouin in Urbanisme II, no. 10, 1933, pp. 105 09/10), and numerous palaces survive (G. Zander, Travaux de restauration de monuments historiques en Iran, Rome, 1968). One may conclude that in Safavid times secular and shrine architecture absorbed the energies of architects to an unprecedented degree, and to the detriment of other religious architecture.

The rate of production was markedly patchy. Periods of intense activity alternate with protracted lulls. As noted above, Safavid rule brought increasing centralization and tended to confine major architecture within the orbit of the court. When royal patronage failed, architecture itself languished, for the period produced virtually no vigorous local schools. The continuous rhythm of distinct local traditions that had underpinned the great architecture of earlier periods was gone. Safavid Iran, so rich in ordinary buildings, is comparatively barren of great ones. Interestingly enough, the obvious exception, the years from 1590 to 1630, is also a period which saw a permanent concentration—indeed, a constellation—of talent in one place. It could very well be argued that for the rest of the century Isfahan lived off the resources accumulated in that creative period, and that these resources created the continuity necessary for further work in the same style.

The intense focus on ornament partially explains why the Safavid architect rejected outright innovation, preferring to refine the relationship between the constituent parts of his building (e.g., palaces of Isfahan). The forms he used were at once traditional and few in number. Now, effortlessly deployed on a gigantic scale, they achieved their fullest maturity. Thus the Masjed-e Šayḵ Loṭfallāh and the Masjed-e Šāh repeat the millennial schemes of the domed square and the four ayvān plan. Such forms become modular and therefore the same schema turns up in mausoleums (Shiraz: Ḵātūn-e Qīāmat and Bībī Doḵtarān, see Survey of Persian Art, figs. 417 18), palaces (Hašt Behešt; P. Coste, Monuments Modernes de la Perse, Paris, 1867) and even on bridges (Ḵᵛājū Bridge, Isfahan). Mosques, caravanserais and madrasas are hard to tell apart (complexes of Ganj ʿAlī Khan, Kermān and Mādar-e Šāh, Isfahan). Religious shrines and palaces alike have display chambers for precious ceramics (Ardabīl and ʿAlī Qāpū).

In secular architecture the emphases are naturally somewhat different, but some of the underlying concerns already discussed, especially the interest in scale and spatial diversity, may also be detected here. Palaces vary from full scale gardens garnished with buildings (Farahābād; Survey of Persian Art, fig. 517) to individual kiosks (Bayrāmābād, near Kermān; ibid., pl. 510B). Figural tiltwork is important (I. Luschey Schmeisser in AMI, N.F. 9, 1976, l0, 1977) for Qazvīn and Isfahan, while in a palace at Nāʾīn the decoration is of thinly incised plaster and comprises poetical quotations, and figural scenes, reminiscent of contemporary painting, rugs and textiles, dominate (idem, “Der Wand und Deckenschmuck eines safavidischen Palastes in Nāyīn,” AMI, N.F. 2, 1969, pp. 183ff.). These palaces deliberately exploit insubstantiality. They minimize obviously solid bearing walls, preferring to pierce the surface by large windows, niches or loggias and to encrust interior walls with yet more niches as well as false low hung ceilings and stalactite vaults. Wood here plays an important structural role which it is accorded nowhere else in major Iranian Islamic architecture. Ground plans and elevations hark back with surprising fidelity to pre Islamic models, notably in such features as the tālār, wooden ceilings, and figural sculpture (e.g., Čehel Sotūn).

Among the varied utility structures of the period (bridges, bazaars, cisterns, dams, pigeon towers, and others) caravanserais dominate. They are starkly functional in character, but despite a generic similarity in layout, encouraged perhaps by the widespread adoption of an official blueprint plan, numerous minor variations are common. The basic vocabulary is simple—vaults, courtyards, vestibules, arcades, and small scale domes—but the range of plans includes octagonal and circular examples as well as the ubiquitous four ayvān type as at Bīsotūn (W. Kleiss “Das safavidische Karawanserei von Bisutum,” AMI, N.F. 3, 1970, pp. 289 308). Popular tradition is probably right to attribute most of them to the reign of Shah ʿAbbās, whose concern to foster trade and communications also led to the construction of the “Stone Carpet,” the great causeway through the Caspian marshes.

Ambitious patronage then, had generated ambitious architecture. But the grand style was lost after the death of Shah ʿAbbās (1038/1629), and the apparent failure of nerve was due rather to the patron than to the architect. In consequence the Safavids achieved qualitatively less in two centuries than did the Saljuqs or the Il khanids in their best fifty years. The sheer scale of projects like the Ganj ʿAlī Khan complex at Kermān (M. E. Bāstānī Pārīzī “L'ensemble de Ganj Ali Xân à Kerman,” in Proceedings of the VIth International Congress of Iranian Art and Archaeology, Tehran, 1976, pp. 13 20) and the redevelopment of the shrines at Ardabīl, Māhān, and Mašhad—to say nothing of the great ensembles at Isfahan—must have entailed multiple unfamiliar problems of logistics and administration. Buildings on this scale were a serious drain on resources and could not lightly be undertaken. Nor was money the only problem. The time scale adopted for some of these buildings was probably inadequate for them to be completed successfully—hence the oft told and possibly apocryphal tale of the architect of the Masjed-e Šāh, who against the royal will suspended work on the building for some years to allow the foundations to settle. Things were very different in the more highly organized building industry of the Ottoman empire.

Stylistic development. The salient features of the Safavid style are best illustrated in the context of specific examples. A consistent emphasis on sheer size climaxed with the Masjed-e Šāh and the Mašhad shrine, respectively the largest mosque and shrine in Iran, and encouraged a rethinking of architectural problems. Radial symmetry is deployed in the grand manner (Farahābād gardens; cf. D. N. Wilber, Persian Gardens and Garden Pavilions, Rutland, 1962, pp. 126 27) or on a more human scale (Hašt Behešt palace; Ḵᵛāja Rabīʿ tomb). Where radial symmetry was undesirable or impractical, the symmetrical repetition of key elements in pairs was often popular (e.g., the constituent elements of the Masjed-e Šāh; cf. L. B. Golombek, “Anatomy of a Mosque: The Masjid i Shāh of Iṣfahān,” Iranian Civilisation, ed. C. Adams, Montreal, 1972, pp. 60 69).

The emphasis on size also encouraged architects to compose their buildings in terms of blocks and independent of applied ornament. This led to a simplifying trend that could result in somewhat banal architecture, but the better Safavid buildings make a direct and satisfyingly strong aesthetic impact. Pools, ayvāns, niches, domes, tālārs, and open spaces are manipulated with remarkable confidence (tomb of Moḥammad Mahrūq, Nīšāpūr; Hašt Behešt; mosques adjoining the Isfahan maydān). Increased scale fostered a new inventiveness in exploring spatial values. When existing buildings were enlarged, whether by courtyards (e.g., Ardabīl shrine), galleries or other features, they were opened up or given stronger axiality; alternatively their several parts were made interdependent. New vistas were created. Sudden contrasts of scale or of lighting created novel interactions of large and small, open and closed spaces (ʿAlī Qāpū and Čehel Sotūn palaces, Isfahan). The visitor to the Masjed-e Šāh (completed 1047/1638) enters a low, shadowy vestibule and his eyes have barely adjusted themselves to this dimness before he suddenly finds himself once more bathed in bright light in the mosque’s resplendent inner courtyard (see axonometric drawing in A. Welch, Shah ʿAbbas and the Arts of Isfahan, New York, 1973). In the Šayḵ Loṭfallāh mosque (completed 1027/1618) these harsh contrasts of lighting are rejected but the contrast of scale between the vestibule and the mosque proper remains.

Tilework is perhaps the key to Safavid architecture. At its best, this tilework almost rivaled earlier achievements, and it was certainly used on a larger scale than hitherto and for new purposes (e.g., dome exteriors). Some combinations of brick and tile were also new (dome of Šayḵ Loṭfallāh mosque). But in general its technique was inferior in that large square underglaze painted tiles of dulled colors replaced the laborious tile mosaic of earlier times, in which each color had been fired at optimum temperature. This permitted huge areas to be covered with tilework quickly and cheaply. Buildings came to be clad in tiles from top to bottom, to the detriment of structural values. As a result Safavid buildings are sometimes dismissed as facade architecture.

Seen as a whole, Safavid architecture seems safe rather than daring; it prefers to simplify rather than to explore complexities. In this sense it is less original than earlier schools of Islamic architecture in Iran. It lacks the internal development of Ottoman and Mughal architecture, being too firmly rooted in the past to encourage innovation. Nevertheless a hybrid Perso Armenian style was fashioned at Julfa, the Armenian suburb of Isfahan, in the 11th/17th century. Here forms derived from Persian mosques and mausolea are translated into an Armenian idiom (J. Carswell, New Julfa: The Armenian Churches and Other Buildings, Oxford, 1968). Architecturally the most novel feature of the period was the confident marshaling of huge spaces. Thus the southern approach to Isfahan was via the bridge of Allāhverdī Khan, some 300 meters long, and the mile long Čahār Bāḡ, an avenue lined with trees, streams, and buildings, while the center of the new city was marked by a great maydān or square, whose area of twenty acres far outstrips European plazas. In the field of polychrome decoration, the continued emphasis on tilework brought a new status to the specialist in its various techniques. The names of craftsmen and calligraphers now proliferate. But the primacy accorded to ornament implied a waning interest in architecture per se.

It is often said that the Madrasa ye Mādar-e Šāh (completed 1126/l714; see Honarfar, Ganjīna) was the last great building to be erected in Persia. Certainly Safavid architects had wrung the best out of traditional forms; understandably their successors used these forms with somewhat less conviction and gradually introduced new elements of European origin. Of Nāder Shah’s reign little survives apart from his mausoleum (at Kalāt-e Nāderī), a gigantic pastiche of a Saljuq polygonal tomb tower with engaged columns and carved stone orthostats of appropriately Mughal style (D. Wright in The Illustrated London News 250, 24 June, 1967). Contemporary mosques at Kermān, Rasht, and Qom survive and Nāder added a gilded dome and minaret to the Mašhad shrine.

Zand architecture in later 18th century Shiraz was fundamentally eclectic (ʿA. Sāmī, Shiraz, tr. N. Sharp, Shiraz, 1958). The city plan apes Safavid Isfahan, the arg has bastions with decorative brickwork of Saljuq type, tiles and bas reliefs revive Sasanian iconography, turrets on ayvāns reflect Indian models and floral compositions on tilework display a distracting realism perhaps borrowed from Europe.

The next major patron of architecture, Fatḥ ʿAlī Shah Qājār, founded numerous spacious four ayvān mosques—at Qazīn, Borūjerd, Zanjān, Tehran, and Semnān—impartially named Masjed-e Šāh. Their size naturally recalls Safavid work; indeed, the Qajars expanded Safavid buildings (e.g., Qom and Māhān) or imitated their decoration, sometimes quite shamelessly (e.g., golden ayvāns or domes at Mašhad, Ray, and Qom). Safavid precedents underlie the remodeled bazaars of Qazvīn and Kermān, with their numerous caravanserais, or multiple foundations like the Ebrāhīm Ḵān complex at Kermān (A. M. Hutt and L. Harrow, Iran II, London, 1977).

In the sphere of town planning, the Čahār Bāḡ-e Ḵᵛājū in Isfahan depends on Safavid inspiration, as do numerous palaces at Isfahan and around Tehran (Coste, Monuments modernes). The scale is often colossal: the Golestān palace in Tehran is a city within a city and the Nīāvarān palace, also in Tehran, had some fifty isolated pavilions, while only 18th century India can match the eight superposed terraces of massed tiny arcades in the Bāḡ-e Taḵt, Shiraz. Immemorially ancient forms were preferred, notably the tālār or columned portico (Dīvān ḵāna in the Golestān; Sarhangābād) and the openplan domed kiosk (Došān Tappa). Garden settings remain standard for these palaces. Easily the finest surviving combination of gardens and architecture, where buildings are embowered in greenery and reflected in pools and canals, is the originally Safavid Bāḡ-e Fīn near Kāšān (Wilber, Gardens).

If palace architecture was traditional, other forms were new (R. Hillenbrand, “The Role of Tradition in Qajar Religious Architecture,” in C. E. Bosworth and R. Hillenbrand, eds., Qajar Iran. Political, Social and Cultural Change 1800 1925. Studies Presented to Professor L. P. Elwell Sutton, Edinburgh, 1984, pp. 352 82). They included the ingenious split level Masjed-e Āḡā Bozorg in Kāšān, with its deep sunken courtyard (H. Narāqī, Āṯār-e tārīḵī-e Kāšān wa Naṭanz, Tehran, 1348 Š./1969) and the exaggeratedly bulbous shape of certain Shirazi domes (Šāh-e Čerāḡ and Sayyed Mīr Moḥammad shrines). Religious architecture languished about 1266/1850, producing little beyond the huge and derivative Sepahsālār mosque madrasa in Tehran, built towards the end of the 19th century. Flamboyantly original, however, and quintessentially Qajar are the decorative gateways used as entrances to bazaars (Yazd), military installations (Semnān), and cities (Qazvīn, Tehran; see E. Pākravān, Vieux Teheran, Tehran, 1962). Here, as in shrines (Qazvīn, Qom, Māhān) minarets proliferate as decorative accessories (Hutt and Harrow, Iran II). French inspired military architecture achieved brief prominence under the Qajars (Tehran; Maydān-e Tūp ḵāna) but it was principally in traditional and vernacular architecture like caravanserais, bazaars, wind towers, and ḥammāms that high standards of construction were maintained.

Qajar decoration is usually unmistakable. Simple, rather strident tiled geometric or epigraphic designs in small glazed bricks were especially popular. The repertory of cuerda seca tiles now included episodes from the epic and legendary past, portraits of Europeans, scenes from modern life, and the country’s heraldic blazon of the lion and the sun (J. M. Scarce in Oriental Art, N. S. 12, 1976). Pavilions and palaces bore figural paintings which revived Sasanian royal iconography (Negārestān palace, Tehran) or betrayed the influence of European illustrated magazines or painted postcards depicting landscapes and tourist spots (A. A. Bakhtiar and R. Hillenbrand, “Domestic Architecture in Nineteenth Century Iran: the Manzil i Sartîp Sidihî Near Isfahan,” in C. E. Bosworth and R. Hillenbrand, eds., Qajar Iran. Political, Social and Cultural Change 1800 1925. Studies Presented to Professor G. P. Elwe1l Sutton, Edinburgh, 1984, pp. 383 401). Bastardized European architectural forms—steeply pitched roofs, decorative fenestration, classical capitals, pediments, rounded arches—combine incongruously with local architectural vocabulary. European styles and themes infiltrate the carved figural stucco of this period (many houses in Kāšān; Hutt and Harrow, Iran II). Also of European origin is the most spectacular Qajar decorative technique—mirrorwork (see ĀʾĪNA KĀRĪ). Reflecting glass now complemented polychrome tilework, adding play of light to play of color (Golestān palace; Hall of Mirrors; Mašhad shrine). The facetted surface of moqarnas vaults was the ideal vehicle for this late but still novel expression of a classic preoccupation of Iranian architecture—the dissolution of surface by resplendent ornament.

Bibliography: Given in the text.

(R. Hillenbrand)

Originally Published: December 15, 1986

Last Updated: August 11, 2011

This article is available in print.
Vol. II, Fasc. 4, pp. 345-349