HAVING now considered the Tâj Mahall as a typical example of Indian design produced under Muhammadan auspices, let us go back to the beginnings of Musulmân rule in India and attempt to realise the peculiar conditions which led to the development of the different styles of architecture usually described as Indo-Saracenic. The classification adopted by Fergusson in his history is most misleading to the student, because, for the purpose of an academic analysis, he has detached all Muhammadan architecture from its historical context, and treated it as an importation unto India of a new order of architecture by an artistically superior race, rather than as a continuous development of Indian building traditions proceeding from altered conditions of social and political life, changes in religious ritual and symbolism, and in the structural requirements evolved therefrom.

The oft-quoted phrase that “the Pathâns built like Titans and finished like goldsmiths” conveys an historical fallacy. The Pathâns were fighting men, not builders; the building traditions they brought with them into India, called Pathân by Fergusson, were those which Mahmûd of Ghaznî and his descendants had borrowed from India. These traditions in the course of two centuries had been adapted to the needs of a militant race. Western writers exalt the simple dignity and grandeur of the Pathân tombs in Northern India, to the disparagement of Hindu temple architecture, without pausing to consider that both belong to the Indian building tradition, and that to draw comparisons between their respective architectural merits is like discussing together the different styles of a Norman keep and a Gothic cathedral. Among the fighting clans of Afghanistan a saint’s or warrior’s tomb on a hilltop was more often a fortress than a holy shrine, and for a reasonable architectural analogy one must put the tombs of the Pathâns in India by the side of the stately Hindu fortresses of Chitor or of Gwalior, or the fort of Agra built by Akbar’s Hindu architects. It will then be easy to understand that the Pathân tombs are as truly Indian as the military works of the Hindus.

The only satisfactory method of studying the Indian building styles is to adopt a chronological basis for the general classification, in the same way as European styles are usually designated by the centuries to which they belong, using provincial or local names to distinguish different subdivisions. When one thus compares a fourteenth-century Indian mosque in Gujerat with Hindu temples of about the same period and locality, it will be evident at a glance that there is no real connection, from an architectural point of view, between the former and Muhammadan buildings in Egypt, Arabia, or Persia, and that the term Saracenic can only be used in a conventional sense, for the mosque and the temple are both Hindu.

The beginning of the thirteenth century, or nearly two centuries after the death of Mahmûd of Ghaznî, saw one Muhammadan dynasty established on Indian soil at Delhi, and another in Bengal at the old Hindu capital at Gaur. The few monuments of these two dynasties which are now extant are either mosques or tombs, which show very clearly that the Muhammadan invaders did not trouble themselves with spreading any new architectural propaganda in India.

The armies of Islâm brought few masons and other craftsmen with them, so the Delhi Sultans and their satraps in Bengal did as Mahmûd had done they impressed the Hindu builders and craftsmen into their service. They wanted mosques for the true believer to be built quickly and magnificently. Mathurâ and other places which had furnished Mahmûd with builders for his capital were in the vicinity of Delhi. The Muhammadans were thoroughly practical in their methods, ‘and, though they hated the idolater, had no scruples against using the splendid materials provided by Hindu temples, and doubtless found a grim satisfaction in compelling thousands of Hindu craftsmen to wreck their own holy shrines and to rebuild them according to the ritual of Islâm.

The building styles of this part of India, which were lithic developments of the early Indian wooden styles, lent themselves easily to the purposes of the Muhammadan iconoclasts. Nothing was easier than to transport piecemeal the splendidly carved columns, with their bracketed capitals and lintels, of the Jain and Hindu temples, and to re-erect them on a plan dictated by the mullahs who superintended the construction of the mosques, which, according to Muslim tradition, consisted of a quadrangle with its two longer walls generally pointing in the direction of Mecca. On the side opposite to the principal entrance was placed the lîwân, or sanctuary containing the mihrâb and the mimbar, or pulpit. The three remaining sides were usually enclosed by narrow colonnades or corridors. The lîwân was necessarily much more spacious than these corridors, and the roofing of it thus presented many more constructive difficulties.

The domes of the Hindu temple mandapas, or porches, supplied ready-made roofs both for the corridors and for the lîwâns of the mosques. Of course the heavy external masonry of the Hindu domes with its elaborate sculptured symbolism was neither necessary nor desirable for the roofing of the mosques. All that was essential for Muslim practical purposes was to take the constructive parts, or the inner stone shell of the Hindu domes (Plate IX), cement them on the outside to make them water-tight, and finish them with the wonderfully fine plaster which Indian masons had used from time immemorial as a preservative for brickwork and as a ground for painted decoration.

No doubt Mahmûd’s Indian masons had followed a similar method in roofing the mosques and palaces at Ghaznî, though in this case they were not reconstructing ancient domes but building new ones. This was the origin of the so-called Pathân or Muhammadan dome in India. It was only a simplified Hindu dome, stripped of its external decoration, but constructed entirely according to Buddhist-Hindu methods. We will discuss these methods later on.

This makeshift mosque, put together by Hindu craftsmen and made decent and proper according to the Puritan sentiment of Islâm by the mutilation of the Hindu figure-sculpture, satisfied for a time all the needs of the faithful. The rapidity as well as economy with which official requirements were provided for by these peremptory and drastic measures might well be envied by our Anglo-Indian administrators. According to tradition, the great mosque at Ajmîr, finished in the reign of Altamsh (1211-35) was put together in two and a half days! Due allowance must be made for Oriental hyperbole, but if the walls of the quadrangle and the arches in front of the lîwân are left out of account, such a performance, with many thousands of skilled craftsmen at command and finished materials already collected at the spot, would not be altogether incredible. The enclosed quadrangle was probably used for prayer, and thus was regarded as a complete mosque, before the roofing of the colonnades was finished.

The methods of the Delhi Public Works Department in the thirteenth century, if more brutal than those of the present day, were decidedly more practical and efficient, not because the Muhammadan military officers and mullahs were superior in architectural taste to the British subalterns and military chaplains, and their coadjutors the British engineer and bricklayer, who have been deputed in these latter days to instruct the despised Hindu craftsmen, but because the Delhi Sultans did not expect their officials to play the part of amateur builders. They were there to rule and enjoy themselves, and to make the Hindus work for them. Teaching the Hindus Saracenic “orders” of architecture did not enter into their official code; they only required that the heads of the faithful at prayer should be protected from the dripping of rain through a leaking roof. The Hindus were acknowledged to be the best builders that Asia could provide, and Islâm had no professional or commercial interests to promote at the expense of Indian art and craft.

The advantage to the Hindus was that, provided that they did not offend the religious susceptibilities of their masters, they were left free to exercise their wits in the practice of their art and craft, and were not subjected to a slow process of intellectual starvation by being put to copy paper patterns provided by official experts not trained in practical craftsmanship and without knowledge of or sympathy for Oriental art traditions. The advantage to Islâm, from a proselytising point of view, was that very many Hindu craftsmen, some from conviction and some from motives of self-interest, adopted the creed of their masters, and thus in process of time a new style of Indian building more perfectly adapted to Muhammadan needs and taste was evolved.

The mosques constructed in the fashion described from the ruins of Hindu temples became the prototypes of others constructed by Indian Muhammadan builders, but it was soon felt that the open colonnades of the corridors and sanctuary afforded too little protection from sun and rain. To remedy this, a screen of brick, sometimes plastered, sometimes faced with stone, was built in front of them (Plate X), and naturally enough the mullahs insisted that the pointed arch, with its symbolic associations for Islâm, should be used for this screen, the only original constructive work in most early Indian mosques, for even the enclosing walls of the quadrangle were originally the walls of a Hindu or Buddhist temple courtyard. The screen served also a ritualistic purpose: instead of symbolic sculpture, the laws of Islâm or sacred texts were carved upon it for the instruction of the congregation. Now, the Hindu masons were quite familiar with the pointed arch as a symbolical and ornamental feature1—from the early days of Mahâyâna Buddhism it had been used in Buddhist and Hindu sculpture—but either from experience of earthquakes or for other practical reasons they mistrusted “the arch which never sleeps” as a structural device, except for very small spans. And since they had generally at their disposal unlimited quantities of first-rate material, either wood or stone, admirably adapted for their traditional beam-and-bracket system of construction, there was no practical reason for using any other; so even when put to building arches of wide span for the Muhammadan mullahs, they made many attempts to adapt their own system to this innovation.

Fergusson’s dictum regarding the great range of arches in the screen-wall of the mosque of Qutbu-d-Dîn—that “the Afghan conquerors had a tolerably distinct idea that pointed arches were the true form for architectural openings”—seems to be founded on a complete misconception, both from an historical and an architectural point of view. It is highly improbable that the Musulmâns who directed the building of the mosque—assuming them to have been Afghans, which is not at all likely—were influenced by any æsthetic reflections, intuitive or otherwise, in insisting that arched openings should be put into the screen-walls. They wanted arches because they were the symbols of their religion. We may assume that they showed the Hindu craftsmen illuminated copies of the Quran or paintings of Arabian and Persian mosques as a guide, but otherwise left them to construct the screens as they pleased. The “Saracenic” arch is not intrinsically more true for architectural openings, either in a constructive or æsthetic sense, than the round arch or the Hindu beam and bracket. These different constructive methods have each their respective advantages. A true craftsman, guided only by practical considerations, would make the choice of any one of them depend upon the character of the opening, its size and position in the constructive scheme, and upon the character and quality of the materials he was using. The Hindu craftsman had very good constructive reasons for preferring the beam and bracket for buildings adapted for his own religious ritual. In the buildings he made for Muhammadans the pointed arch may have added to his constructive resources, but it was in no sense scientifically superior to his habitual methods of construction. Indeed, modern developments of building construction, in which iron is so largely employed, reduces the pointed arch to the place it generally held in the Hindu system, namely, to a decorative expedient only, and makes the beam and bracket of the Hindus the scientific form of construction. For this reason, if for no other, the Hindu building craft is worthy of more attention than it has yet received from the Anglo-Indian departmental expert.

The very ruinous state of the mosque at Old Delhi makes it less interesting as an architectural example than the almost contemporary building at Ajmîr, built in the same fashion. The original mosque in the courtyard of which stands the famous iron pillar, 2 a wonderful monument to the scientific knowledge and skill of Hindu craftsmen many centuries before the Muhammadan invasions was commenced by the first Delhi Sultan about 1196; the screen of arches in front of the lîwân were added by his successor Qutbu-d-Dîn about ten years later. Altamsh, the next Sultan, who succeeded in 1210, began to enlarge the mosque by extending the lîwân with its screen north and south, and by adding a great quadrangle which should have enclosed the original building. The next Sultan, ’Alau-d-Dîn (1296-1316), built a fine gateway on the south side of this outer quadrangle, and projected yet further extensions of the building which were never completed, and the present mosque is only a fragment of the original, for nearly the whole of the lîwân behind the arches and a considerable part of the corridors surrounding the two quadrangles have disappeared. The great Tower of Victory, in what remains of the outer quadrangle, known as the Qutb Minâr (Plate XI), built about the same time as the original mosque, belongs to a class of monument in which the Hindus excelled; though this one is a Saracenic modification of the Indian type, of which the two towers at Chitor are the best extant examples. They were no doubt derived from Buddhist structures, which again may have had their prototypes in Babylonia and Assyria. The three finely proportioned lower stories of the Qutb Minâr, which were probably designed by masons from Ghaznî, belong to the original tower; their exceeding beauty is greatly marred by the upper part, which is a badly conceived restoration and addition of the Sultan Firuz Shah (1351-88). A “classical” cupola added to the summit by a Public Works engineer in the early part of the nineteenth century has fortunately been removed.

Though used as a place from which the mu-azzin should summon the faithful to prayer, the tower of Qutbu-d-Dîn3 has no connection architecturally with the adjacent mosque. The two minarets of the latter were comparatively insignificant and placed on either side of the great central arch of the screen of the lîwân, more for ornamental than practical purposes. Only small fragments of the two minarets on the Ajmîr screen now exist. In later buildings, in which they become much more important, both structurally and ornamentally, they were frequently removed to the extreme ends of the screen of a mosque, or placed at the four corners of a mausoleum. In the Tâj we find them detached from the building and placed at the four corners of the platform on which it stands.

The most important contribution of Saracenic art to the Indian building craft of the thirteenth century was not constructive but decorative. Some of the Arabian mullahs were past masters in calligraphy, and in the beautiful Kufic and Tughra script the quotations from the Quran carved on the screens of the mosques (Plate X) made magnificent decoration and admirable sermons in stone. Fergusson admits that in carrying out this work the Indian craftsmen excelled their teachers. “As examples of surface decoration,” he justly observes, “these two mosques of Altamsh at Delhi and Ajmîr are probably unrivalled. Nothing in Cairo or in Persia is so exquisite in detail, and nothing in Spain or Syria can approach them for beauty of surface decoration.” But when the same high authority proceeds to discriminate between “Muhammadan largeness of conception” and “Hindu delicacy of ornamentation,” one must question his judgment in drawing such a distinction between the Musulmân and Hindu artistic genius. The remains of the magnificent Hindu architectural works constructed before and during the time of the first Muhammadan invasions of India prove that largeness of conception was no monopoly of the Saracenic building tradition; and as the earliest Muhammadan buildings in India were undoubtedly built almost entirely by Indians, and mainly according to their own ideas, we should give full credit to the infinite skill and versatility of the Indian builder, who, with an unbroken craft tradition of many centuries behind him, could and did adapt it as perfectly to the formula of the Muhammadan mullah as to that of Buddhists, Jains, or Brahmans.

It may seem to the Western eye, trained in the formula of the classical schoolmaster, that the Muhammadan prescription is more pleasing, just because it is more correct according to the canons called classical; but the creative impulse in the great art produced in India under Muhammadan rule, which seems to us so admirable, belonged to the same Indian races and the same Indian civilisation and culture which had inspired the works of earlier times. If the Indian craftsman of to-day is often a mere copyist, it is chiefly because the methods of our teaching and the principles of our administration have made him so. The whole of Muhammadan architecture in India bears this distinctive impress of the soil to which it belongs—that its structural ideas and symbolism are nearly always essentially Indian, not foreign importations: the foreign suggestions adopted by Indian builders were almost without exception purely decorative ones—e.g. the use of Persian and Arabian floral and geometric motifs for surface decoration in place of Hindu sculpture, and the substitution of encaustic and painted tiles for painted plaster or terra-cotta.

These foreign borrowings were never mere copies, but were always given a distinctive Indian character, even when they played an important part in the decorative scheme of a building, just as in European art the frequent adaptation of Oriental ideas can generally be recognised as European.

The planning of Muhammadan buildings, the arrangement of the interior, the various forms of the roofing and its supports, whether columns, piers, brackets, pendentives, or arches, were almost invariably derivations from Buddhist or Hindu craft traditions. The screens of pointed arches often make an Indian mosque appear Saracenic from the outside; but directly one enters, it is evident that the building is as much Indian as a Hindu temple.

A comparison of Muhammadan buildings of the thirteenths century in India with one which was being constructed in the same century on the Western extremity of the vast territory then under Musulmân rule may be useful for showing how little India really borrowed from Saracenic sources. The Alhambra of Grenada is one of the most typical and famous of Saracenic buildings. Here Arab civilisation, instead of adapting the building traditions of conquered races to its own purposes, was almost for the first time trying to create something which should be wholly after its own ideals. The Moors of Spain tried to cast off the traditions of Rome and Byzantium, of Egypt, Persia, and India, which had hitherto helped them and other Muhammadan races to make magnificent monuments for themselves: they would show what the genius of Islâm could create for itself out of brick and stone. The result may be called magnificent as decoration, but it was not building—rather stage architecture suggestive of gorgeous scenery inspired by illuminated Arab manuscripts, and often made constructively absurd by the painting, gilding, and stucco.

In the thirteenth-century mosques of Delhi and Ajmîr it is evident that the Arabian calligraphist and painter had their say with regard to the decoration, but the craftsmanship, both decoratively and constructively, was Indian, and fine because it was Indian. The construction of the arches was according to the Hindu bracket system: the weakness which manifested itself in some of them after many centuries is not due to a faulty system, but to the fact that, like the Egyptians of old, the Muhammadan taskmasters expected their captives to build with unsuitable materials, i.e. with stones too small for the Hindu method of bridging over open spaces in walls.

  • 1. When Buddhist or Hindu niches containing the images were large, they were sometimes vaulted, so that the arch became structural as well as decorative.
  • 2. It is attributed to the time of the famous Hindu King Vikramaditya, who flourished in the fifth century A.D.
  • 3. Mr. R. N. Munshi, in his History of the Kutb Minar (Bombay, 1911), gives reasons for attributing it to the reign of Qutbu-d-Dîn’s son-in-law and successor the Sultan Altamsh, who also built the mosque at Ajmîr.