Delhi

Front matter

FLaNS and sections

REPORT FOR THE YEAR 1871-72.

A. C. L. CARLLEYLE,

Preface

following report Mr. Beglar admits that the pillars have been more or less re-arranged, hut he contends that they occupy their original positions in the colonnade of a single Hindu temple, and that their present height is exactly that of the original Hindu colonnade. Consistently with this view he is obliged to condemn the record of the Muhammadan builder of the masjid regarding the destruction of twenty-seven Hindu temples as a false boast.

This opinion I consider as quite indefensible. The Muhammadan conqueror could have no possible object in publishing a false statement of the number of temples destroyed, nor in recording a lie over the entrance gateway of his great masjid. I therefore accept the statement as rigidly true. It is besides amply confirmed by the made up pillars of the colonnades on three sides of the court, which, as I have shown in my account of Delhi, must certainly have belonged to a great number of different temples.

That the Kutb Masjid was the site of a single Hindu temple I have no doubt whatever ; and before Mr. Beglar began any excavations under my instructions, I stated to him my opinion that the lower portion of the surrounding walls of the raised terraee on which the masjid stands was the original undisturbed platform of a Hindu temple. The excavations made by Mr. Beglar outside these walls have proved that my opinion was cori'ect. I therefore look upon this raised terrace on which the Kutb Masjid stands as the site of the chief temple of the old Hindu city of Delhi, of which the only remains now existing in silu are the tall Hindu pillars immediately behind the great arch. Tliese are true Hindu pillars, undisturbed and unaltered by the Muhammadans. I consider them undisturbed, because their places are marked out on the pavement by chiselled lines; and I call them unaltered, because they have single Hindu shafts like those of other Hindu temples, whilst all the pillars in the surrounding colonnades are Muhammadan compositions, made up of two or three separate pillars to obtain height. I can offer one proof as to the pillars behind the great arch being in their original positions, in the fact that the stones of the piers of the great arch are actually cut out so as to fit against the mouldings of the pillars. This fact seems to me to show that the arches were an after addition built by the Muhammadans against the pillars of the old Hindu temples, which, therefore, I conclude to be in situ.

PREFACE.

iii

In the cloisters this curious mode of fitting new work to old work is found exactly reversed in the case of one of the pilasters of the northern colonnade. Here a natural bulge on one of the wall stones is met by a corresponding hollow cut out of the hack of the pilaster. I call them pilasters because they are placed against the walls ; but they are frequently full pillars with one face turned to the wall, and their bases not being reduced to bring their centres under the midlines of the shafts, these bases are projected several inches out of their true positions beyond the central lines of the shafts. I conclude, therefore, with absolute certainty that the pillars of the surrounding colonnades are not in situ, but were added by the Muhammadans to the old Hindu walls of the courtyard of the ruined temple.

As a further proof of the patch-work character of these incongruous pillars, I can adduce the following simple facts : —

1. In the north cloister, the first pillar in the outer row to the east has an almost plain granite shaft for the lower member, which is placed upside down.

2. In the north cloister, two contiguous columns of the outer line are each formed of three pieces of similar pillars placed one above the other, with the same mouldings and the same ornaments. Of these six pieces four are octagons with the alternate faces indented, the other two being plain octagons ; but in one the plain octagonal piece forms the top third, in the other it is the middle third.

3. Several plain octagonal shafts, as well as others with the alternate faces indented, are found resting on square blocks, which were originally intended to receive square shafts. This is clearly shown by the upper surfaces of the exposed corners of the square blocks, which are not covered by the present octagonal shafts, being still in a rough state, just as they left the mason’s hand, furrowed with chisel marks.

But the proofs of re-arrangement of old materials are not confined to the pillars of the colonnades. They are equally numerous and equally convincing in the domes of the gateways and comer rooms of the cloisters. These domes Mr. Beglar considers to be in their original positions, hut the following facts will show that they also have been re-arranged by the Muhammadans : —

1. In the south-east comer the dome springs from an octagon which rests on a square supported by eight pillars,

JV

PREFACE.

namely, four corner pillars and four middle pillars. Tlie angles of the octagon, therefore, fall on the beams of the square instead of over the pillars.

2. In the dome of the north-west corner this notable fault is intensified by the spaces between the pillars being unequal, as the intermediate pillars are not placed in the middle of the sides of the square.

3. In the north-cast corner the dome springs from an octagon supported on twelve pillars forming a square ; hut there are capitals of five brackets placed in the corners of the square, and some common capitals of four brackets in the angles of the octagon. How did these brackets get into the wrong places except by re-aiTangcment ?

4. In the great dome of the cloister of the east gateway there is the same kind of anomaly ; some of the beams of the octagon resting in the angles between two brackets of common four bracket capitals, instead of being placed on the angular bracket of a five bracket capital. The beams also are of different kinds, some being plain. and some ornamented. This dome must, therefore, have been re-arranged in its present position by the Muhammadans.

5. In the smaller ceilings of the side spaces of the corner rooms, there are some square ornamented slabs with their sides cut to make them fit into oblong spaces, whilst othera are uncut, but are eked out at the ends of the oblong by plain slabs to fill the blank spaces. Tliese roofs, therefore, are certainly not in their original positions, and I conclude, without any hesitation, that they must have been arranged as they are now by the Muhammadans when the great masjid was erected by Kutb-ud-din out of the spoils of twenty-seven Hindu temples.

6. In the upper rooms of the north-cast and south-east comers many of the roofing slabs appear to have been taken from a Jain temple, as the faces of several of the stones which reduce the size of the square openings by covering the angles are filled with figures of men, elephants, and horses, with a single squatted figure in the middle, quite naked, and with both hands lying in the lap, exactly after the fashion of Jain statues. If then, as Mr. JJeglar argues, these cloisters are the colonnades of a single Hindu temple, that temple must have been dedicated to the Jain worship, a conclusion which is directly negatived by the inscription on the iron pillar standing in the midst of the coui’t-yard.

preface.

V in Avhicli the pillar itself is called the “arm of Vishnu.” It is also at variance with the several Vaishnava sculptures which are huilt into the surrounding walls, such as the Ten Avat&rs, and Nhrtlyan reposing on the folds of the snake Ananta.

The conclusion which I have come to regarding the Kutb Masjid is simply this, — that it was iJViilt by Hindu masons under Muhammadan supervision out of the ruins of twenty-seven Hindu temples, some of w hich were no doubt Jain. This would be quite sufficient to account for all the faults of construction which have been noted, as well as for the incongruous make up of the pillars of the colonnades. The object of the Muhammadans was to obtain height, and as the generality of Hindu shafts are short monoliths, they gained their object by building up a single tall column of two or three Hindu shafts piled one over the other, with portions of bases and other blocks interposed. The general effect is no doubt pleasing, hut I believe this to be due solely to the profusion of ornament, which distracts the eye, and prevents it from observing the incongruity of thick shafts surmounting thin ones, of bracket projections supporting nothing, and of niche projections unsupported. If the pillars were plain, I believe that their utter want of symmetry would be signally striking, and the incongi'uity and W'ant of harmony of the several pieces would be obvious at once.

The Hindu origin of the Kutb Minar is also upheld by Mr. Beglar, who argues in favour of his opinion with much ingenuity. The Hindus themselves claim the pillar as their own, and assert that it was erected by Prithi llAj to enable his daughter to see the Ganges ! In a former report I have stated my reasons at length* for believing that the Kutb Minar is a purely Muhammadan building, and I will now adduce others which have occurred to me during two separate visits to the Kutb Minar in company with Mr. Beglar. On the first visit he pointed out to me the following short inscriptions in Nagari letters recorded on the lower part of the minar : —

1. — On plinth outside, to right of entrance —

2. — On wall of passage inside door to left —

* Arclwological Surrey of Icdu, rul. I, p. 190.

VI

PREFACE.

3. — On underside of lowest overlapping archstone in entrance passage —

Samrat 1256.

I take these three inscriptions to be the records of one of the Hindu masons employed in building the minar. That they are productions of an illiterate person is shown by the omission of the letter'® in the first, and of the figure 1 for 1,000 in the second record. These three records of the same date Samvat 1256, or A. D. 1199, seem to me to point either to the foundation or to the completion of the building in that year — the repetition of the date being a common practice amongst Hindu masons. Thus, on the pillars of the AtAla Masjid at Jonpur, which is known to have been originally a Hindu temple converted to Muhammadan use by Ibrahim Shah Sharki between the years 1403 and 1440 A. D., I found the date of Samvat 1464, or A. D. 1407, three times repeated as follows : —

1. — On right jamb of north-gate outside —

Samrat 1464 Samapt

Sutrad/idro Padumavi

Sat Su/radAdfa Suta.

“ Finished in the Samvat year 1464 by the mason Padumavi, son of the mason Sai”

2. — On one of the lower square pillars —

Samvat 1464.

Buniddipari.

" Founded in the Samvat year 1464.”

3. — On one of the outer pillars on the south side —

Samvat 1464.

But in addition to these dates, I can cite an actual record of the master mason, or builder who superintended the erection of the minar. This is cut on the south face of the plinth, and was first pointed out to me by Mr. Beglar. The record is unfortunately rendered imperfect by the fracture of the stone ; hut the remaining letters and figures are sufficiently distinct —

% X Ma gaj 51i 83J dftranftmum

Immediately to the right of the figures 61^ there is a broad arrow, or long upright lino with a stroke sloping downwards on each side — and about one inch to the right there is another similar mark which has been partially obliterated. Both of these lines are continued on the upper surface of the plinth. The words ddrandmuni I take to mean the “ exemplar line,” or as we should say the “plumb-

PBEFACE.

VU line” of the building, which the shifting of the mark shows to have been readjusted to the extent of rather more than one inch at the time that the record was made. That this was done by a Hindu mason is proved by the adoption of the word gaj, instead of the Persian gaz, which would certainly have been used by any one of the Muhammadan conquerors. It is curious that I found a similar mark on the projecting part of the back wall of the masjid itself immediately behind the centre, which was obviously intended for the adjustment of the mid-line of the masjid itself, as the mark is several inches outside of the middle of the projection.

Mr. Beglar bases his opinion chiefly on two arguments : —

1st . — That there is a difference of style observahlo between the three lower storeys and the two upper storeys, from which he infers that the former must be Hindu work and the latter Muhammadan.*

2nd . — That the distanees between the bands of ornamentation are in geometrical progression ; from which he argues that as the chain of the series must have been a work of “ no ordinary labour,” it could not have been diseovered by the “ barbarous Muhammadan conquerors,” and must therefore be the work of the intellectual Hindus. t

In his first argument Mr. Beglar ignores the fact that the two upper storeys were re-built by Firoz Tughlak, as recorded in the inscriptions, as well as in his autobiography. The difference of style is indeed very striking, but it is not necessary to suppose that it marks more than the difference of architectural taste which had taken place in a century and a half that had elapsed between the times of Kutb-ud-din Aibeg and Firoz. The rich style of ornamentation of the lower balconies is in strict accordance with the lavish traceries of the arches which is seen in both the early masjids of Delhi and Ajmer, while the contrast of white marble with red sandstone was a favourite device in the time of Firoz, as may still bo seen in his pillar at Hisar. That the style of ornamentation is Hindu may be admitted; but this was a simple necessity of the early Muhammadan architecture of India, as the conquerors were soldiers who were naturally obliged to employ the masons of the country in carrying out their designs. Hence came the overlapping arches as well as the Hindu ornament.

pup;kack.

Mr. Boglar’s second argument, which depends on the use of a recondite geometrical aeries, I utterly repudiate as a mere fanciful guess. Even supposing it were true, it is not easy to helieve that the armies of Muhammad Ghori were formed entirely of illiterate men when we remember that the learned Abu llih&n accompanied the still earlier expeditions* of Mahmud of Ghazni. But when it is admitted that “ the chain of the series (geometrical) must alone have been a work of no ordinary labour,” I must confess that I put no faith in the application of such a practically impossible scries.

Mr. Beglar further argues that there is a difference of projection between the hands of lotus flowers and the bands of Arabic inscriptions, the former being flush with the faces of the column, while the latter have a considerable projection. But it is only the lowest hand of writing that is flush with the face, all the others being raised ; and this belt has been so much injured by time and by ignorant restorations that it is now illegible, Sayid Ahmed being able to read no more than the words Amir ul TJmra, or “ chief of the nobles.” I believe therefore that the depression of this belt is due entirely to the restorers, a conclusion which is borne out by the fact that several of the red facing stones have been cut right through so as to leave the inner core of rough stone quite visible. Now, this belt forms no part of Mr. Beglar’s geometrical series, and as I am quite satisfied that it was an integral part of the original ornamentation, I repudiate the whole scheme of recondite Hindu design as a mere fanciful theory.

But on this subject of the ornamentation of the Kulb Minar we now possess the most decisive evidence that it is not of Hindu origin in the T&rikh-i-Alai of Amir Khusru, a contemporary of Alauddin Khilji. Speaking of the new minar, which this king had ordered to he built, the poet adds that he also “ directed that a neio casing and cupola should he added to the old one.”t According to this account, which we must remember is that of an eye witness of the fact, the whole of the present red stone facing was added by Alauddin, and to him we must assign all its exquisite balconies as well as all its ornamental bands. It was no doubt

f Tirikli-i-Aliii iu Dowiiou's edition of Sir II. Elliot’s Muhammadan Historians, Jll, 70.

PREFACE.

IX the knowledge of this entire restoration by the Patlian King that led Baber to call it the minar of Alauddin Khilji.

Having thus disposed of the Hindu origin of the ornamental bands on the face of the pillar, I will add a few words about the minar itself. In proof of the Hindu origin of the column, Sayid Ahmed states that “ there is only one minar, which is contrary to the practice of the Muhammadans, who always give ttco minars to their masjids.” But this statement is correct only for the custom of the last three centuries, as the following facts will show that the previous jiractice of the Muhammadans was to have only one minar, or mazinah, to their masjids : —

1. — The masjid of Ibu TAlhn in Cairo, built in A. D. 876, has only one minar.*

2, 3. — The two minars of Mahmud at Ghazni, built about A. D. 1000, are of different sizes, and stand half a mile apart. They therefore belonged to two different masjids.

4. — The masjid of Sultan BarkAt in Cairo, built in

A. D. 1149, has only one minar.f

5. — The minar at Koel, built in A. D. 1252, was a

single column, and occupied exactly the same position with regard to its masjid as that of the Kutb Minar to its masjid.

6. —The unfinished minar of Alauddin at Delhi is a

single column built about A. D. 1300.

7. — Two masjids at Bayana have only one minar each,

placed outside the court-yard of the masjid, but on the north-east corner instead of on the south-east corner as in the Kutb example. One of these bears an inscription of Naseruddin Muhammad, who was reigning in A. D. 1390.

From these eight examples, which range over a period of more than five centuries down to within one hundred and sixty years of the accession of Akbar, it is clear that it was the fixed practice of the Muhammadans to have only one minar, or mazinah, to each masjid. That the Kutb Minar was a mazinah or Muazzin’s tower we have the evidence of its inscriptions which cannot be ignored, as well as the express declaration of Abulfeda who calls it a mazinah. That it was a Muhammadan design we have the fact, that

• Hintorv of ArehiU'cturr, FergHRSon I]» 383, 387. t Ibid. II. 387.

TREFACE.

A

the conquerors of Delhi were the rulers of Ghazni, where they had been familiar with the minars of MahmAd, whieh are star polygons in plan, with deeply indented angles. For the Muhammadan the Kutb Minar had a purpose which was intimately connected with the daily praetiee of his religion. I am therefore firmly convinced that it is a purely Muhammadan building both in purpose and in design, although most, if not all, of the details of its execution, and notably its overlapping or corbelled arches, are Hindu.

AGRA.

With regard to the buildings at Agra there are only two points in which my opinion differs from those which are generally received by the public, and which have been adopted by Mr. Carlleyle. According to the popular view the present fort was built entirely by Akbar, there being little if any part of the old fort of Sikandar Ludi included within its limits, and the red sandstone palace now called the Jahhngiri mahal was built by the emperor Jah&ngir for his own residence. As I do not agree with either of these opinions, I will now give the reasons which have led me to differ from them, and as the main part of my argument about the fort rests upon the true determination of the site of Jah&ngir’s palace, I will begin with the latter building.

The emperor Jah&ngir, on returning to Agra a few years after his accession, relates that he first visited his father’s tomb, and then proceeded to the fort “ to the palace which he had ordered to be built there for his own residence.”* That this palace, which he had ordered to be built for his own residence, was not the red sandstone building now called the Jahdngiri mahal, is clear from the emperor’s account of its position : “ This palace,” he says, “ rests upon the gate which opens on the river Jumna, and is supported by 25 pillars.” The actual palace of Jahhngir must therefore have been quite close to the water gate of the fort towards the well known Musamman Buij, or octagonal tower, as he adds that “ the adjoining tower of four storeys is octagonal.” Now, this is an exact description of the Musamman Burj, and I conclude therefore that the palace of Jahhugir must have occupied that portion of the eastern face of the fort lying between the water gale and the Musamman Buij which is

• Aatobiogrnphy of Jahingir, p. 120,

PREFACE. xi

now represented by the fine court-yard of the Macchhi Bhawan, backed by the Diw&n-i-khtls. Here still stands Jah^lngir’s famous throne of black stone, which was seen and described by Captain Hawkins in A. D. 1611 and 1612.* The Diw&n-i-khfts with its two thrones of black and white stone is now a terrace, 70 feet wide by 120 feet in length ; but as Shah Jahan’s Shish mahal has cut off about one-third of its length, the original terrace must have been nearly 200 feet in length with its back wall towards the river, and its front open towards the great square of the Macchhi Bhawan. Against this wall to the right and left of the throne were placed images of Christ and the Virgin, which are described by William Finch, as well as by Captain Hawkins. Their accounts have since been confirmed by the discovery of a white marble statue of the Virgin, which was dug up in the fort about 16 years ago, and may now be seen in the verandah of the Agra Bank.

On this terrace then, as I believe, stood the original palace of JalAngir, supported, as he describes, on 25 pillars, the whole of which must have been cleared away by Shah Jahan before he began to build the well known Shish mahal and Kh&s mahal.

The emperor also adds that “ adjoining is a small gallery, overlooking the Jumna, to view wild boast fights this agrees with the account of Captain Hawkins, who regularly attended Jahilngir’s darbars for nearly a whole year, and who repeatedly witnessed wild beast fights as well as executions from the gallery of the palace. As there is no gallery attached to the red sandstone palace, it seems certain that this building cannot have been the palace of Jah&ngir.

The emperor closes his account with the characteristic entry that in a lower storey, “ more on a level with the river,” was the “ private drinking apartment.” Now, the building called the Jahfingiri mahal has no apartments below the audience halls, while immediately adjoining the Macchhi Bhawan, that is, between it and the water gate, there are two octagonal rooms, in a lower storey, each 21 J feet in diameter, which were once highly decorated with painted flowers, and which still possess two large marble trellises looking towards the river. Here again the Macchhi Bhawan buildings agree with the description which Jahdngir gives of

Digitized by Googic

I’liKKACK.

xii

his own |)alace, Avliilo the building now called the Jahiingiri raahal has nothing in common with the emperor’s account either in position or in details. I conclude, therefore, without any hesitation that the red sandstone palace now called the Jah&ngiri mahal was not the palace of Jah&ngir.

But if this bo not the palace of Jah&ngir, to whom must it be attributed ? I reply to Ibrahim Ludi. It is certain that there was a palace of Ibrahim within the fort of Agra, as Baber records that he occupied it on his arrival at Agra, and afterwards notices that there was an empty space between Ibrahim’s palace and the ramparts.*

In this “ empty space” Baber constructed a large wain (or Baoli) 20 gaz in diameter. This wain I take to be the great Baoli, 17^ feet in diameter, which still exists in the empty space between the red sandstone palace and the ramparts. But the argument to be derived from an examination of the building itself is even stronger and more conclusive than this identity of position. This argument lies in the simple fact that the red stone palace of Agra is a direct imitation both in design and in detail of the M&n Mandar or palace of Baja M&n Sinh at Gwalior. Now, Ibrahim was the conqueror of Gwalior, and the fine palace of Eaja M&n must certainly have won his admiration, as it did that of Baber only a few years later. Gwalior was captured in A. D. 1518, and Ibrahim fell on the fatal field of PAnipat in 1526. The palace, therefore, must have been built during the eight years comprised between these two dates.

The main points of agreement in construction are the following : —

1. The entire absence of radiating arches, all openings being spanned by corbels.

2. The use of sloping stone ribs meeting horizontal beams to carry the slab roofing of the two halls of audience.

3. The adoption of narrow screened galleries running round the whole building, and of secret passages in the thickness of the walls for communication between the upper and lower storeys.

4. The adoption of a narrow gallery screened close by trellis work round three sides of one of the audience halls, apparently for the purpose of allowing the ladies of the harem to see into the great court-yard.

PHKKACK.

xiii

The chief points of similarity in the details of ornamentation are the following ; —

1. The use of running borders filled with ducks and pigeons.

2. The style of the stone trellises which is the same in both palaces.

To all these proofs of an earlier date than the time of Jah&ngir, I may add the fact that the walls arc covered with delicate diaper patterns of small and intricate tracery, having the ground or sunken portions coloured either with a vermilion or a sky-blue pigment, — a style which, according to my experience, was peculiar to the later Pathan architecture.

With all these concurring evidences pointing to an earlier date than the time of the Mughals for the building of the so-called Jahdngiri raahal, I have been led Irresistibly to the conclusion that the Ked Palace was erected by the last Pathan King of Delhi, and that it is therefore the actual palace of Ibrahim Ludi which was inhabited by the victorious Baber.

After I had come to this conclusion I determined on visiting the palace once more to inspect the building called a Bdradari, which I have always heard attributed to the Pathans, and which must therefore have formed part of the old Pathan Fort known as Bddalgarh.

On entering the fort by the Amar Singhi Gate, I was struck by the marked difference of styles in the two kinds of battlements which crown the middle gateway. On the outer face they are plain, massive, and lofty, with clear straight openings, like those of the fort walls; but on both sides as well as on the inner face the battlements are low with raised border.s, and slightly recessed openings widening below into indented squares, which were once filled either with glazed tiles, or with some blue or green colour. All the buildings in the court-yard between the middle and inner gates are finished with the same battlements. So also are all the buildings in the court-yards of the Delhi and water gates, while the outer battlements correspond with those of the walls. Now these smaller battlements are precisely the same in all respects as those of the Bed Palace called the Jahftngiri mahal. If then I am right in attributing the Red Palace to Ibrahim Ludi, it is only a necessary conclusion that the main portions of the three principal entrances to the fort,

XIV

PUEFACE.

including their inner court-yards, must have formed parts of the original fort of tlie Ludi Kings.

This conclusion is further borne out by the use of diapered medallions on the walls and inside the domes, a practice which was generally given up by the Mughals except for the ornamentation of the spandrils of arches. This conclusion is also much strengthened by the fact that the old B&radari, which I have always heard attributed to the Pathans, has precisely the same battlements in all respects as those of the Red Palace and the inner courts of the gateways.