Pramāṇa · Search · About · API

The Geography of the Puranas c7.s5: Puṣkaradvīpa

The Geography of the Purāṇas c7.s5

Language
English

Puṣkaradvīpa

c7.s5.1

The accounts of this dvīpa are given in different Purāṇas at considerable length. The Brahmāṇḍa Purāṇa says that there is no rainfall, no springs, no fountains in this dvīpa nor is there any vegetation. The Skanda says that between the two provinces of Puṣkaradvīpa (Mahāvitam and Dhātake Khaṇḍam) runs the mountain Mānasa in a circle (like a full moon-Matsya). The account in the Varāha and Viṣṇu is much the same. The note in the Brahmāṇḍa Purāṇa about the mountain chains being one in two or two in one appears also in the Liṅga and Kūrma Purāṇas. All the Purāṇas agree that the exterior province is Mahāvitam and the interior is Dhātaki. It is stated in the Matsya Purāṇa that the Mahāvitam is the western portion and exterior to the mountain. The Vāmana Purāṇa and the Garuḍa Purāṇa give a harrowing account of the prevailing climatic conditions and dwell on the fiendish savagery of the tribes living in it. It says, 'The dvīpa is a land of horrors, devoid of purity, cruel and leading to the destruction of the soul. It is the land of demons, full of awful hells twentyone in number'. Other Purāṇas, however, mention Puṣkara dvīpa as a paradise (Viṣṇu) which conflicts with the text of the above descriptions.

c7.s5.2

The descriptions given above can be reconciled if we note all the facts mentioned in the account of Puṣkaradvīpa and analyse them with the help of a relief map of the world (Fig. 3). These points are:

c7.s5.3

(1) Puṣkaradvīpa is situated in the same latitudes as Uttara Kuru. Since the seasons there are the same as in the latter (Matsya and Vāyu), we must search for Puṣkaradvīpa in northern latitudes.

c7.s5.4

(2) The climate of Dhātaki and Mahāvitam which form part of the Puṣkaradvīpa is of 'dry' type, since 'It does not rain there, neither are there any rivers in those regions. The climate is also not influenced by heat or cold (i.e., it is dry all the year round). The water supply and plant life of the country depends on water from the hills' (Matsya).

c7.s5.5

(3) The name of the dvīpa is derived from the fact 'it is surrounded by Puṣkara (lakes of lotuses)* (Matsya). Therefore we have to find a land in the northern latitudes in the vicinity of which there are a number of lakes.

c7.s5.6

(4) There is, according to Matsya, a huge circular mountain chain named Citrasānu situated in the eastern half of the dwīpa. The western half is dominated by another great circular range named Mānasa with Mahāvita (son of Mānas) as its spur covering its western rim. Other Purāṇas also agree that there is a mountain range running through the whole extent of the dwīpa and dividing it into two provinces one of which is exterior and the other interior with respect of the above mentioned ranges. In addition some Purāṇas qualify this statement by saying that this chain of mountains is really a continuous range but appears as split into two, meaning thereby that there is a saddle or break in the range which allows access across it between the exterior and interior region. Mānasa, according to Matsya, looks like a full moon rising near the sea coast.

c7.s5.7

(5) A marked feature of the dwīpa is its occupational structure. The practice of rearing domestic animals, trade, husbandry are non-existent there* (Matsya). This points to the fact that the people living there are nomads and hunters and in general primitive and savage as described in other Purāṇas as well.

c7.s5.8

(6) These are two contrasting pictures of the dwīpa, a paradise on the one hand and a desolate waste on the other. This signifies that the ‘central range is fairly high and is a distinct climatic barrier between its constituent parts. One side is a dry desert and the other, suitable for human occupation. It promises a paradise for those who approach the dwīpa from one direction, while it presents the appearance of a wasteland if one enters it from the opposite direction. Such phenomena of knife-edged boundaries between two regions of strong contrast are not uncommon. For example, if we approach Peru from the west, it presents a dry, treeless, barren appearance, while if we enter it from the east it presents the most luxuriant equatorial forests.

c7.s5.9

(7) It is surrounded by an ocean of fresh water and surrounds the sea of milk (Kṣīra-Sāgara).

c7.s5.10

The regions which broadly satisfy conditions (1), (3), (5) and (7) above are the Scandinavian lands, Finland, Northern European Russia and Siberia, since these lands occur in the same latitudes as those of Uttara Kuru. They contain numerous lakes, support nomadic people who live by hunting and are washed by the Arctic waters or the waters of the cold current which flows past the Bering Sea along the eastern coast of Siberia.

c7.s5.11

to Japan. These waters have the lowest salinity among the oceans of the world and may be called 'fresh water'.

c7.s5.12

Condition (2) excludes Scandinavia since its rainfall is between 40" and 80", much more than what is required by this clause. In fact, Scandinavia is one of the fairly moist regions of the world, which is contrary to the picture of the western segment of Puṣkaradvīpa presented by the Purāṇas.

c7.s5.13

Condition (4) excludes Finland and North European Russia. In the latter case the Urals may be taken as one of the ranges of Puṣkaradvīpa but in the first instance it is not circular (a point on which all Purāṇas are unanimous) and secondly, there is no other mountain range which could represent the other equally important ranges which occur on its east or west.

c7.s5.14

The western portion of Siberia is included in Uttarakuru, a sub-region of Jambū Dvīpa by the Purāṇas.⁴⁴ The central and eastern parts of Siberia are the only two regions which satisfy conditions (1) to (5) and (7). Of these two, only the eastern part of Siberia and the adjoining lands satisfy (6). This narrows down our investigation to only one region—the north-eastern seaboard of Asia which qualifies for identification as Puṣkaradvīpa. Let us examine this region in detail concentrating only on inhabited or habitable lands which are covered by this region.

c7.s5.15

The Sea of Japan is enclosed by a ring of mountain ranges. On its east are the Japanese Highlands and on its west are the Highlands of Korea, Manchurian Coast (culminating in Peishan peak) and Sikhota Alin which 'rise near the sea coast'.⁴⁵ This circular mountain chain is almost continuous with very narrow breaks in the Korea strait and the Gulf of Tartary. Suppose we call the Japanese Highlands, Chitrasanu, the eastern segment of the circular range of Puṣkaradvīpa and the western coastal ranges as Mānasa, the picture of Puṣkaradvīpa appears to emerge on the world map. If this assumption is true we should have another range Mahāvīta parallel to Mānasa to its west. There happens to be a range, the Khingan range which lies to the west and almost 'guards the western frontiers' of this region.⁴⁶ To the east of this range which may be called Mahāvīta, there is a prosperous and rich land, which corresponds to the central plains of Manchuria while a desert Dhātaki Khaṇḍa on its west is represented the Gobi desert in our scheme.

c7.s5.16

If these premises are correct, then the mountain Pitrsoma is the mountain Fujiyama⁴⁷ of Japan, the Kṣīra Sāgara is the Sea of Japan and the Sea of Fresh Water is the north-western Pacific. Puṣkaradvīpa, therefore, covered the whole of Japan, Manchuria and the south-eastern Siberia.

c7.s5.17

*Śālmaladvīpa*

c7.s5.18

(1) The characteristic tree of this dwīpa is *śālmalī* or silk-cotton tree which gives the dwīpa its name. This tree is commonly found on the margins of equatorial regions or of monsoon lands with moderate rainfall. Hence we should locate Śālmaladvīpa in the rainy tropical belt of the old world.

c7.s5.19

(2) The annual range of temperature in this dwīpa is low, since according to Matsya Purāṇa (122, 93-99) it has neither summers nor winters, that is, there are no seasonal variations there.

c7.s5.20

(3) It is a region of high cloudiness since, 'no stars, planets or the moon are visible'.

c7.s5.21

(4) The people of this dwīpa obtain the best victuals of all the six flavours without any effort on their part (Matsya, 122, 100). This brings out the significant characteristic of the region that the people who live there are 'food gatherers' and not 'food producers'. The vegetation cover produces enough food to satisfy their needs.

c7.s5.22

The above features of Śālmaladvīpa are more than enough to justify its location in the equatorial region of the old world. There are two such regions within this belt, the equatorial lands of Africa and East Indies. The latter were included traditionally in the South-Eastern Peninsulas of Jambu Dwīpa.⁴⁵ Therefore, by the process of elimination we reach the conclusion that by 'Śālmalidwīpa', the Purāṇas mean the tropical part of Africa bordering the Indian Ocean on the west. It included Madagascar—the Zenj of the Arab and Persian geographers the Harina of the Purāṇics, and the Sankh Dwīpa of some other writers.

c7.s5.23

⁴⁵ [Note text missing from page]

c7.s5.24

*Krauñcadvīpa*

c7.s5.25

It is a pity that the Purāṇas do not mention any climatic or vegetational characteristics of this region. Therefore we have to fall back upon miscellaneous information regarding the dwīpa contained in not only the Purāṇas but also in other works. The information collected from the various sources is preseted below:

c7.s5.26

(1) Mount Krauñca from which the dwīpa derives its name, appears for the first time in Taittiriya Āraṇyaka (1, 31, 2). This means that it has had associations with Jambū Dwīpa (and incidentally with India) since remote antiquity. In all probability, therefore, this region was situated in the immediate vicinity of Jambū Dwīpa.

c7.s5.27

(2) The Mahābhārata (12, 14, 21-5) locates Krauñcadwīpa in the west (of Meru) while the same epic (6, 12) places it in the north. The inevitable conclusion is that the dwīpa touched the Jambū Dwīpa in the north-west. The Bṛhat Saṃhitā (14, 24) and the Rāmāyaṇa (4, 43, 25) locate it in the north.48

c7.s5.28

(3) The dwīpa is watered by ‘thousands of streams’49 in addition to the seven important rivers ‘flowing with great volume of water’50 mentioned by name. The dwīpa, therefore, is definitely a humid region with abundant rainfall.

c7.s5.29

(4) Although all the dwīpas, except the Jambū, find different places in the sequence of dwīpas enumerated by different Purāṇas and other works, it is noteworthy that Kuśa and Krauñca always appear successively in all of them. It suggests that in all probability the two dwīpas were situated close to each other.

c7.s5.30

(5) The statement that ‘the Krauñcadwīpa surrounds the Ghṛitoda sea like the tyre of a wheel’51 provides an important clue to the location of this dwīpa.

c7.s5.31

We have therefore to find an inland sea situated to the north-west of Jambū Dwīpa, virtually surrounded by a well-watered populated52 belt containing a number of mountains which stand in or near it and also a number of voluminous rivers which flow through it.

c7.s5.32

There are five enclosed or partially enclosed seas to the north-west of Jambū Dwīpa, the Baltic, the Mediterranean, the Black Sea, the Caspian and the Aral Sea. Of these only the Black sea practically fulfils all the conditions mentioned in 1-6 above. The Baltic, the Aral and the Caspian are flanked by deserts (hot or cold) which are either uninhabited or uninhabitable and therefore the bordering lands do not qualify for the term ‘dwīpa’ applied by the Purāṇas. Besides their characteristics as regards climate, rivers and mountains are just opposite to those mentioned in the Purāṇas. The Mediterranean region is too large a unit to be considered as an annular region surrounding an inland sea. Besides, its southern margin is a desert practically reaching its southern shore We may, therefore, conclude that the Krauñcadwīpa of the Purāṇas is represented by the basin of the Black Sea.53

c7.s5.33

CHAPTER III The Mountain System of the Puranas

c7.s5.34

MERU or the abode of Brahmā is the pivot of and the key to the Purāṇic geography of the world. It is the point of reference round which are symmetrically arranged the mountain systems of the Purāṇic world. The mountains delimit and are closely associated with the regions and the countries, which collectively form the Purāṇic continents.

c7.s5.35

The Vedas do not mention Meru. It is referred to as Mahāmeru or the ‘Great Meru’ in the Taittirīya Āraṇyaka (1,7. 1,3), but there is no indication of its location and size. The Epics, the Buddhist and the Jaina scriptures and the Purāṇas contain practically identical notices regarding the size and extent of Meru and its central location in relation to the continents or major regions of the world. The Persians, the Greeks, the Chinese, the Jews and the Arabs repeat the traditional nodality of the Meru in their cosmography. Although the details and denominations vary, the central idea is practically the same as in the Purāṇas. The Bundesh and Vendidad Sada maintain that the central continent, Khonneretz Bami is surrounded in the east, south and west by a sea called Zare-Ferakh-Kand and in the north by a high mountain named Al-Burj which corresponds to the Purāṇic Meru. The Al-Burj according to Zoroastrian books not only gave birth to all the mountains of the world but it was from this mountain, that the streams which water our earth descended.¹ The Chinese too believed that the centre of the earth from south to north lay in the high and snow-covered Tsing-ling mountains (or Sumeru), from which emerged chains of high mountains and which gave rise to many rivers.² The Jews, as well, made this mountain a part of their cosmographical system. Isaiah, making use of such ideas as were generally accepted in his times, introduced Lucifer (eq. Swarbhanu in Sanskrit) boasting that he would exalt his throne above the stars of God and would sit on the mountain of the Congregation (cf. ‘Sabhā’ in Sanskrit), ‘on the side of the north’. Meru, according to Hindu tradition, is also the rendezvous of gods where the Sabhā or the assembly of gods is held on its northern side. The Mountain of God is also alluded to in the psalms, though, in some instances, it seems to imply Mount Moriah.

c7.s5.36

In the Purāṇas the descriptions of Meru, its location and extent and the great mountain-ranges associated with it as well as the regions which are bounded by these ranges and their sub-divisions are very much similar. The Agni and Brahma Purāṇas are word for word the same as the Viṣṇu and the Kurma, Matsya, Mārkaṇḍeya, and Vāyu present many passages common to them and the Viṣṇu, or to one another. The Vāyu, as usual goes more fully into details. The Bhāgavata differs in its nomenclature of subordinate details from all, and is followed by the Padma. The other Purāṇas either omit the subject or refer to it only very briefly. The Mahābhārata (Bhīṣma Parva) presents an account which is essentially the same, and many of the stanzas are common to it and the different Purāṇas.

c7.s5.37

Regarding the size and shape of Meru, the Viṣṇu Purāṇa says, ‘Jambū Dwīpa is in the centre of all these (continents); and in the centre of this continent (Jambū Dwīpa) is the golden mountain Meru. The height of Meru is 84,000 yojanas, and its depth below the surface of the earth is 16,000 yojanas. Its diameter at the summit is 32,000 yojanas, and at its base, 16,000: so that this mountain is like the seedcup of the “lotus of the earth”.’¹

c7.s5.38

The shape of Meru, according to the above description, is that of an inverted cone; and on the analogy of the seedcup, its form should be circular: but there seems to be some uncertainty regarding this subject amongst the Purāṇics. The Padma compares its form to the bell-shape flower of ‘Dhaturā’. The Vāyu represents it as having four sides of different colours, i.e., white on the east, yellow on the south, black on the west, and red on the north. There are other opinions as well regarding the outline of the mountain Meru, e.g., according to Atri it has 100 angles, to Bhṛgu, a thousand; Sāvarṇi calls it octangular, Bhaguri, quadrangular and Vārṣāyani says it has a thousand angles; Galava makes it saucer-shaped; Garga, twisted like braided hair, and others maintain that it is circular. The Liṅga makes its eastern face the colour of ruby; its southern, that of lotus; its western, golden; and its northern, coral. The Matsya has the same colour as the Vāyu and both contain this line, ‘Four-coloured, golden, four-cornered, lofty’, but the Vāyu compares its summit in one place to a saucer, and observes that its circumference must be thrice its diameter. The Matsya also, rather incompatibly, says the measurement is that a circular form, but it is considered quadrangular.

c7.s5.39

According to the Buddhists of Ceylon, Meru is said to be of the same diameter throughout (i.e. cylindrical). Those of Nepal conceive it to be shaped like a drum. The description of Meru and its surrounding mountains, as given in the Brahmāṇḍa is exactly the same as in the Vāyu.

c7.s5.40

The above references bring out the following information regarding Meru:

c7.s5.41

THE PURĀṆIC DWĪPAS

c7.s5.42

(1) Meru signifies a vast plateau of immense height lying in the heart of the central continent, Jambū Dwīpa. Its height and size, relative to other mountain systems and regions, is shown in a sketch (see Fig. 4). This sketch has been drawn to scale, according to the description given in the Viṣṇu, Matsya, Vāyu and Bhāgawata Purāṇas.

c7.s5.43

(2) It indicates quite clearly that Meru is not the name of an isolated peak, a single mountain, or an individual range. It is a fairly extensive plateau hemmed in by a ring of mountain peaks, so that the top of Meru resembles a saucer—i.e., it is depressed in the centre and elevated along its margins.

c7.s5.44

(3) The sides of Meru are extremely precipitous. In other words the vast plateau is bounded on all sides by steep escarpments, so that ascent to the top of the plateau from any point on the base is not only difficult but very nearly impossible. The Purāṇas emphasise this steepness by saying very aptly and concisely that the shape of Meru resembles a cylinder. This will happen if the escarpment is just steep. In places where it overhangs it would give the impression of an inverted cone—with a smaller base and a wider top. If the escarpment has a convex profile, it at once brings to the observer's mind the shape of a drum.

c7.s5.45

(4) The assertion that Meru is a circular, square, octagonal or thousand-sided, implies that observers must have travelled round Meru and taken different routes—all close to the escarpment. Those who changed their bearings four times to circumnavigate the Meru, naturally described it as square in shape while those who changed them frequently described it as many sided as the number of times they had to change their general direction of movement. The best way of summarily describing the circumference of Meru was therefore what Viṣṇu Purāṇa says, i.e., it is circular which simply means that one can make a complete round of Meru and come back to his starting point. The shape of a vast massive mountain system like Meru could not be ascertained or visualised otherwise.

c7.s5.46

(5) The height of Meru is given as 84,000 yojanas while that of the ranges north and south of it (including Himavān, Himavat, etc.) is given as 2,000 yojanas (see Fig. 4) which gives the N-S section across Jambū Dwīpa). This appears to be very much out of proportion since there is no mountain in the world which is 42 times the height of the Himavān or Himavat. The inference is clear that the staggering escarpment of Meru when observed from its base, appears to be of tremendous height, even higher than the highest mountain, for the obvious reason that from this position one cannot see the peak or top of the escarpment which dissolves in the misty haze above.

c7.s5.47

Keeping the above facts in mind and assuming that Himvat is the same as our Himālayas, let us attempt to identify Meru in the great mountain complex which lies north of the Himālayas—Himālayas taken in its broadest sense, i.e., the chains of mountains which extend from east to west and bound Bhārata in the north.

c7.s5.48

The accompanying map of the Himālayan and Trans-Himālayan peaks* (Fig. 5) depicts in a simplified form and in a generalised manner the mountain systems of Asia. The peaks of the mountains are classified according to their magnitude and some idea of the relative heights of the different ranges can be visualised at a glance from the map. Besides, a north-south cross-section has also been added to bring out the sequence of ranges and intervening valleys.

c7.s5.49

It will be seen that there are only five elevated regions in the vast Himālayan and Trans-Himālayan zone which could possibly be identified with Meru and whose characteristics approximately, if not exactly, corresponded with those laid down in the Purāṇas and analysed in paras 1-5 above.

c7.s5.50

These are:

c7.s5.51

(a) The region covered by the Karakoram cluster of peaks.
(b) The region covered by Dhaulagiri cluster of peaks.
(c) The region covered by the Everest cluster of peaks.
(d) The Tibetan Plateau enclosed by the Kun-lun and Himālayan arcs.
(e) The Pāmir high plateau enclosed by snow-clad peaks of the Hindū-kush, Karakoram, Tien Shan and Trans-Alai system.

c7.s5.52

The great Karakoram complex lies between the Tarim and the Indus basins, west of the political frontier of Tibet. It lies roughly between 70°E and 79°E, its physical limits being the Hunza-Gilgit valley in the west and the Nubra-Shyok valley in the east, and its northern limit is rather the parapet above the Tarim basin than the Shimshal-Shaksgan depression, while a 'straight' line that can be drawn through Gilgit, Ronda and Khapalu, never above 10,000 feet makes an admirable southern limit. The focus of this complex is the Boltoro glacier, from which apart from K2 (28,250') ten other peaks radiate at heights between 23,000' and 28,000' and no equal area, even of the Himālayas can rival this mass of glacier-mantled giants.

c7.s5.53

Overhanging the Shaksgan valley and within ten miles of the river are three Teram Kangri peaks (24,200'-24,500'), two Gasherbrum (26,000'-26,500') and two sentinels of K2—Broad Peak (26,400') and Staircase Peak (24,750'); and over the western parting, Masherbrum (K1) seems to have two peaks above 25,600' high and Bride Peak which is 25,100' high.

c7.s5.54

Though there is this special focus, three parallel ranges may be distinguished most clearly and most easily as North, Central and South, even if special features or historic names are associated with them, e.g., Aghil, Muztagh and Ladakh.

c7.s5.55

The above description of the Karakoram cluster partly satisfies the conditions which according to the Purāṇas, apply to Meru. It is no doubt a vast, snow-covered, inaccessible, high plateau dotted with numerous magnificent peaks and surrounded by the deep ravines of rivers which flow round it. But the following primary facts do not qualify it for its identification with Meru:

c7.s5.56

(i) The shape of the cluster: it is an oblong region and can in no way be considered or imagined even as approximately circular in shape.

c7.s5.57

(ii) The most important fact that the four principal rivers which flow from Meru to the four cardinal points is not true of this region.

c7.s5.58

(iii) The top of the plateau, taken as whole, is more like that of a corrugated roof (i.e., parallel ridges with irregular intervening valleys) and not 'saucer-shaped' as described by the Purāṇas.

c7.s5.59

(iv) The major regions of Jambū Dwīpa which lie to the East, West, South and North of Meru according to the Purāṇas do not occur as such with reference to this cluster.

c7.s5.60

The Dhaulagiri cluster consists of 6 great peaks (above 24,000') within a radius of about 9 miles, situated on the Great Himalayan Range and its western bifurcation at Dhaulagiri. They are linear summits surmounted by great peaks of almost the same altitude. However, its size, shape, drainage and regional relationship do not in any way conform to the pattern prescribed by the Purāṇas. It may be safely ignored so far as its identification with Meru is concerned. This observation also applies to (c) above, which, with the exception of the fact that they are high plateaus, do not satisfy the other essential conditions laid down in (2) to (5) above. Tibet is too big a unit to be considered as a possible answer to the Purāṇic Meru. Even if we take the whole plateau of Tibet as Meru, the arrangement of mountain ranges and corresponding regions with respect to Meru becomes asymmetric which is against their general alignment as given in the Purāṇas.

c7.s5.61

The plateau of Pāmir, though constantly spoken as such is no more a plateau than the Armenian crown, which it resembles in some essentials; for it is approximately a square segment of the folded belt, running 150 miles both east and west and north and south. But while the mass of the Armenian square is a real crown, from which the land falls in all directions, the Pāmir is really a shallow bowl (cf. the "saucer-shape" of the Purāṇas), overhung by the gable-ends of the Hindūkush, the Tien Shen Kun-lun and the Himālayas. These gable-ends justify its name⁵ (cf. the names: Pa-meer and Meer-u). This Pāmīr bowl (or saucer) is bounded all around by high ranges studded with snow-covered jagged peaks, enveloped in cloud and festooned with glistening glaciers. These ranges are the Alai and the Trans-Alai in the north, the Sarikol and the Kashghar on the east, the Hindūkush on the south and the Khwaja Mohammad range on the west. The peaks which occur round the plateau are, from the north in a clockwise direction, Pk. Stalin (24,623'), Communist Academy Peak (19,239'), Peak Lenīn (23,542'), Peak of Peter I range (above 23,000'), Kungur Peaks (No. 42 and 48, 25,146'), Mustagh Ata (24,388'), the peaks of Wakhan, Shugnan, Rushan and Yazgulem (above 17,000').

c7.s5.62

The Pāmīr is the focal point of Central Asian drainage. Major rivers flow from it towards west (Oxus), south (Indus), east (Tarim) and north (Syr Daria). It is also flanked by E-W mountain chains to the north and south.

c7.s5.63

On these premises, one can safely come to the conclusion that the Meru of the Purāṇas can be identified with the Great Pāmīr Knot of Asia. If we accept the identification of Meru with the Pāmīr plateau, the location of the principal mountain ranges of Jambū Dwīpa as given in the Purāṇas can be investigated.

c7.s5.64

According to the Purāṇas, north of Meru there are, in order, three mountain ranges, the Nīla, Sweta and Śṛṅgavān (see Fig. 6). They are respectively the boundary ranges of Ramayaka, Hiraṇmaya and Kuru, and spread east and west up to the ‘Salt Ocean’. They are two thousand yojanas wide. In length each succeeding range is a little shorter than the preceding one. Similarly, to the south of Meru are the three ranges, the Niṣadha, Hemakūṭ and Himālaya (or Himavān) spreading east and west like the preceding ones. The width of these ranges is the same as that of the northern ranges. They are the boundary ranges of Harivarṣa, Kimpurṣa and Bhārata-Varṣa.

c7.s5.65

Between the Niṣadha and Nīla there are three varṣas. The one round Meru and bounded by Gandhamādana on the west and Mālyavān on the east is Ilāvṛta Varṣa. On the east of Ilāvṛta Varṣa is Bhadravarṣa and on its west is Ketumāla. The relative positions of these ranges and the varṣas which they delimit are shown in the accompanying diagram of Jambū Dwīpa. This diagram does not incorporate one more characteristic of the northernmost and southernmost ranges, i.e., the Śṛṅgavān and Himavat (as given in some Purāṇas, for instance, the Matsya and the Vāyu) that both these ranges are curved like a bow with their concave sides towards the north and the south respectively.

c7.s5.66

A comparison of this diagram with the accompanying physical map of Central Asia will show that:

c7.s5.67

*Śṛṅgavān* represents Kara Tau-Kirghiz-Ketman chain. *Śveta* represents Nura Tau-Turkistan-Atbashi chain. *Nila* represents Zarafshan-Trans-Alai-Tien Shan chain. *Niṣadha* represents Hindukush-Kunlun chain. *Hemakūṭa* represents Ladakh-Kailash-Trans-Himālayan chain. *Himavān* represents the Great Himālayan range.

c7.s5.68

There, however, appears to be a discrepancy. The Purāṇas say that all these ranges extend from sea to sea. This is literally true in the case of the Himavān only. Why and where did the Purāṇics make this obvious mistake in the case of other ranges? If we look at the map of Asia closely we find that the Śṛṅgavān, Śveta and Nila end on the west in the Kizil Kum and Kara Kum deserts, and on the east in the Gobi desert. The Niṣadha extends from the Caspian to the China Sea (taking Tsinling Mountain as its eastern extension). The Hemakūṭa extends from the sea on the east to the Iranian desert on the west. Thus in each case, the range extends from sea to sea—if we define 'sea' as a large expanse of sand or water. Did the Purāṇics hold the view that as barriers to human settlement and movement, the sea of sand and that of water, were synonymous when considering the geography of the inhabited world?

c7.s5.69

Details of a few main ranges, mentioned above, are given in the Purāṇas in different chapters. There are whole chapters on Meru and the Himavān, Niṣadha, Devakūṭa and Kailāsa ranges but the account of the ranges north of Meru is strikingly bare and sketchy. Besides, while the southern ranges are sanctified by their association with a number of gods, the northern ranges are in general not so honoured. The reason is obvious; the Purāṇics were more interested in the southern and central regions of Asia with which they were directly associated for thousands of years.

c7.s5.70

About the Niṣadha, the Purāṇas say⁶:

c7.s5.71

Now I will describe fully the western Niṣadha range. In the central peak of that range is the beautiful temple of Viṣṇu, decorated with gold. It is served by saints and Ṛṣis and is full of Yaksas, Apsaras, Gandharvas, etc. God Hari, the creator in yellow garments, himself resides there and saints are always serving him. In the internal peak of that Niṣadha range is the beautiful city of demons named Ulanghi, decorated with a variety of metals. There are gates of gold and silver and festoons of bright gold and the city is full of passages, caves and roads. Rows of palaces and houses shine like burnished gold and there are beautiful gardens at several places. That city is full of snakes, which guard the

c7.s5.72

entrance gate, and make the entry of enemies impossible. That city of thirty yojanas is always in prosperity. On its southern side several demons reside and in the central part of that range is the city called Deśapradeśa. The entrance to this town is through a 'cave'. On the western peak there are several prosperous cities of Devas, Demons and Nāgas. At the foot of that mountain there is a Somaśilā where the moon shines once a year. Ṛṣis, Kinnaras and Gandharvas always worship the master of stars, Candra. The peak of that mountain is called Brahma-Pārśva Peak. In heaven it is called the abode of Sūryadeva Brahmā. Yakṣas, Gandharvas and Demons serve Mahātmā Brahmā through salutation and worship. There is also the residing place of the fire god Vahni Deva) which is famous in all lokas. Saints, servants, etc. are worshipping Vahni. On the northern beautiful peak of that mountain where Ṛṣis, Saints and a variety of Demons reside, is located a city named Hemacitra, famous in all the three lokas. There are three places of three Devas. On the eastern peak, in the central, and on the western are the places of Nārāyaṇa, Brahmā and Mahādeva respectively. They are always worshipped by powerful Demons, Gandharvas and Yakṣas. On the three peaks of that superior mountain, there are at places beautiful cities of Yakṣas, Gandharvas and Nāgas....

c7.s5.73

The Western Niṣadha is the Hindūkush. Let us study the location of its principal peaks. The highest peak of the Hindūkush is Trichi Mir (T32, height, 25,436'), 10 miles N N E of T32 is another peak T56 (height 24,611') and 9 miles N E of T32 is peak T65 (height 24,543'). On the main range about 25-30 miles towards E is the peak named Sidi-Istragh while further east on the main ridge is another but much lower peak Lankho (22,641'). The last one, though lower in height, is of special importance, since it dominates the route which linked the main E W route across the Pāmir region with the Indian route which branched off from it across the Baroghil pass through Darkot.

c7.s5.74

These are all the high peaks of Hindūkush (or Western Niṣadha). A simple sketch (Fig. 7) showing these peaks and their correct alignment with the main range, would not only enable us to understand the passage given above but would also help us in locating the settlements referred to therein.

c7.s5.75

It would be observed that the Purāṇic location of peaks on the western Niṣadha (Hindūkush) range tallies exactly with the position of peaks as depicted in the sketch. Trichi Mir is the Purāṇic Trikūṭa whose three majestic peaks T32, T56 and T65 represent the Hindu Trinity, i.e. Mahādeva (Śaṅkara), Brahmā and Viṣṇu (Nārāyaṇa) respectively. The main range shown on the sketch and which skirts the Pyandzh in the south is associated with Viṣṇu, though no specific names or associations are indicated for Sidi Istragh or Lankho. The Purāṇas also mention that on the northern and southern slopes and at the foot of the main range lived the barbarians (i.e., 'Demons') while north of Trichi Mir-in the valley of the Oxus and the upper and middle Kokcha valley there were many cities of Devas (or civilised people). These cities of Devas refer to the settlements of Afghan Badakshan which have been mentioned by Holditch¹ in these terms:

c7.s5.76

The northern slopes and plains of Badakshan between the mountains and the Oxus, form part of a region which once represented the wealth of civilization in Asia. The whole region was dotted with towns of importance in medieval times and the fame of its beauty had passed down the ages from the days of Assyria and Greece to those of the destroying Mongol hordes.

c7.s5.77

The barbarous tribes, living near the central peaks of the main range, are the uncouth and savage Kafirs, who, driven from many directions into the most inaccesible pastures of the hills by the advance of stronger races north and south, have occupied remote valleys preserving their own dialects. The settlement of Deśapradeśa lying to the south of the main range probably refers to Dashpar region, which was occupied by the warlike Turkish raiders who were for long a menace to the routes and settlements of the northern valleys. The city of Ulanghi (lit., 'a saddle or pass') is probably associated with the Baroghil Pass and corresponds to Darkot (lit., a 'gateway'). Hemacitra is obviously Chitral, an important town of the N.W. Frontier. The abode or the 'zone of influence' of Vahni Deva points definitely to the settlements of fire-worshippers of Wakhan, Ishkashim, Shignan and Roshan who dominated the upper valley of the Oxus for centuries.

c7.s5.78

It is surprising that the Purāṇic account of the Himālayan chains, i.e. the Hemakūṭa and Himavān is more poetic than precise or convincing. It is a sort of generalised description which runs as follows:

c7.s5.79

Sūtajī remarked: Kailāsa is located on the delightful central peak of Deva-Kūṭa mountain. It is one hundred yojanas in length and fifty yojanas in breadth. Its peak is white, calm, delightful and of good expanse. Several saints live there. There is a big city full of rows of big houses decorated with gold and diamonds. This city belongs to the great god Kubera. That city is safe from the enemies' attack and is magnificent and prosperous. There is a meeting chamber named Vipula in the midst of that town which has several pillars and is decorated with gold in different ways. There is also an aeroplane named 'Puṣpaka' decorated with numerous gems. It contains all essential things, is very fast-moving, is propelled only by desire and is also fitted with gold wires. This aeroplane is used by Mahātma Kubera. Yakṣendra and Piṅgala Deva, worshipped by all Demons reside there along with Yakṣas, 'Gandharvas, Kinnaras, Saints, Panegyrists and Apsarās. Mahātmā Kubera has eight everlasting treasure houses named Padma, Mahāpadma, Makara, Kacchapa, Kumuda, Śaṅkha, Nīla and Nandaka. These are full of jewels. Indra also resides near Kubera, Fire-Yama, and Devas and other Apsarās also reside there. Attendants first worship Mahātmā Yakṣendra and then walk behind him. There is also the Mandākini river with placid water, where ghats are studded with jewels, and beautiful flowers blossom. They are decorated with golden, blue and beryl-coloured lotuses. In that water, the wives of Yakṣas and Gandharvas always bathe and the water of that Mandākini river is drunk by Devas, Demons, Gandharvas, Yakṣas, ghosts and others. Apart from this two other rivers named Alakananda and Nanda also flow there, full of nectar-like water. This water is used by the Devas. On the eastern peak of that mountain there are ten cities of Gandharvas.... There is no limit to the prosperity of these cities.

c7.s5.80

There are, several rows of big palaces. Bright like fire, powerful Harikeśa, Citrasena, Sara, etc., the ten Gandharva kings are the owners of that place. On the western peak of that mountain there are separate palaces of Yakṣas. That place is eighty yojanas in length and thirty yojanas in breadth. There are several big houses. There are thirty prosperous palaces of the people. Happiness always reigns in these houses and thirty Yakṣas like Mahāmāli, Sunetra, Maṇivara, etc., are the owners of those houses. The illustrious Vaiśravaṇa is the master of the aforesaid Yakṣas.

c7.s5.81

On the southern side of that Kailāsa, the Himālaya is located. There are several gardens. There are many falls and caves. There are innumerable peaks, passes and plains. This Himālaya is stretching from the eastern sea to the western sea. At several places on these mountains there are the cities of Kinnaras, which are located in the central part of the many-peaked Himālaya, about one hundred in number. Here men and women are always happy and the subjects are healthy and prosperous. One hundred kings, like Druma, Sugrīva, Sainya, Bhagadatta are the owners of these powerful Kinnaras. In this mountain Mahādeva Rudra was married to Umā. Umā Devi did her hard penance here. Mahādeva in the garb of Kirāta played here and Mahādeva and Pārvatī saw the whole of Jambū Dwīpa. The playground of Rudra Deva is full of different ghosts and full of strange flowers and fruits. Beautiful, big-eyed, happy Kinnari women reside in the caves and mountains of this land of mountains. Here Yakṣas, handsome Gandharvas and innumerable Apsarās are always happy. There is the famous Umāvan. Here Śankara converted his body, half into male and half into female. There is Śaravaṇa where Kārtikeya was born. Here he expressed his desire to destroy Krauñca Sailavana. Kārtikeya has got his chariot here which is adorned with a flag. The destroyer of enemies left his power in the Krauñca mountain full of coloured and strange flowers. Great Devas like Indra made Kārtikeya, the commander of the Devas who was decorated here.

c7.s5.82

He was bright like twelve Suns. There are several places of Kārtikeya, full of ghosts, etc. In the beautiful hinterland of the Himālaya mountain there is Kārtikeya's playground named Pāṇḍu Sila. It is said that on the beautiful eastern land is the residing place of saints which scholars call as Kalapagrāma. In the Himālaya are the āśramas of Mārkaṇḍeya, Vasiṣṭa, Bhārata, Nala, Viśvāmitra, Uddalaka, etc., and hundreds and thousands of āśramas of Ṛṣis. There are several residing places of saints. Yakṣa Gandharvas roam there and several low-class people also live there. That mountain is rich in several minerals. Several types of creatures live there. Thousands of rivers flow from that mountain.*

c7.s5.83

In the opening sentence the term 'Devakūṭa' should be corrected as Hemakūṭa. The Devakūṭa is a different mountain range which with Jaṭhara has been defined by the authors of the Purāṇas as a double range joining the Niṣadha and Nila on the east of Meru. It has, therefore, nothing to do with the Kailāsa mountain. The great range Hemakūṭa is to the north of and parallel to the Himavān. In the above paragraph Kailāsa is supposed to be located on the 'Central Peak' of Devakūṭa. Obviously Devakūṭa here stands for Hemakūṭa.

c7.s5.84

The facts which are brought out by the above description of the Hemakūṭa and Himavān may be summarised as follows:

c7.s5.85

(i) The Hemakūṭa is the modern Kailāsa Trans-Himālayan range.

c7.s5.86

(ii) Purāṇic Kailāsa is only a peak of the Hemakūṭa mountain. It is the existing Kailāsa mountain and not the Kailāsa range.

c7.s5.87

(iii) The traditional city of Kubera (Alakāpurī of Kālidāsa's Meghadūta) lies on or near the Mountain Kailāsa.

c7.s5.88

(iv) The eight 'treasure houses' of Kubera represent eight lesser peaks of the same range which lie in the vicinity of the Kailāsa mountain.

c7.s5.89

(v) The location of Mandākini, Alakanandā and Nandā show that the Purāṇics were thoroughly conversant with the minor topographical features of the Mānsarovara basin.

c7.s5.90

(vi) There were prosperous settlements ('10 cities of Gandharvas ruled by 10 Gandharva kings') on the east of Kailāsa mountain while on its west 'thirty palaces of Yakṣas' were located. This indicates the importance and superiority of the eastern region which had well-settled and better organised population groups over the western region which had a poorer, less civilised and scattered population. It is to be noted that the 'eastern region' recognised correctly by the Purāṇics in the above statement is a geographical region with a high percentage of sunshine, a moist, fertile soil of fine quality, highly successful agriculture, excellent crops of barley, very fair wheat and abundance of vegetables. It rightly deserves its name, Lha-sa or Gods' country.

c7.s5.91

(vii) It is true that the Himālaya is situated south of Kailāsa and that the Himālayas extend from sea to sea.

c7.s5.92

(viii) The Himālayas have a much larger number of peaks, though the Purāṇics say that it has about hundred of them. The names of the peaks are not given here, although a few names occur in other chapters elsewhere.

c7.s5.93

(ix) The inhabitants of the Himālayas are 'Kinnars' (cf. Kinnarā-deśa) and Mahādeva appeared there 'in the garb of a Kirāta.' This indicates people with a lower cultural level dwelling in Himālayan forests, some of which are named as Umā-Vana, Sara-Vana, Krauñca Śaila-Vana, etc. In addition to the forest people there are mentioned innumerable ('hundreds and thousands') āśramas of saints and Ṛṣis in the secluded valleys of this mountain.

c7.s5.94

(x) The Himālayas are rich in minerals and are the source of 'thousands of streams'. Naturally this means that the Purāṇics here refer to the southern face of the Great Himālayan range.

c7.s5.95

Besides the six major ranges, discussed so far, there are a few other ranges which form part of the general pattern of the Purāṇic mountain system of Jambū Dvīpa. These are the Gandhamādana and the Mālyavān which lie to the west and east of Meru and separate the Ilavṛta from the Ketumāla and the Bhadravaṛṣa respectively.

c7.s5.96

The Gandhamādana and the Mālyavān are ranges the location of which is highly controversial. In fact, there is a confusion in the Purāṇas about all the ranges which immediately surrounded or flank Meru. The Viṣṇu Purāṇa says, 'Meru, then is confined between the mountains Nīla and Niṣadha (on the north and south) and between Mālyavān and Gandhamādana (on the west and east)'. The Gandhamādana is also mentioned in other Purāṇas as one of the southern buttresses of Meru (Matsya), as one of the filament mountains on the west, as a range of boundary mountain on the south, and as a Varṣa mountain of Ketumāla (Vāyu). The position of the Mālyavān is also changed in accordance with the position of the Gandhamādana. The Bhāgavata gives different names to the different parts of Gandhamādana, for instance, the buttress is called Merumandāra, the filament mountain, the Haṃsa, while the term Gandhamādana is restricted to the eastern range.

c7.s5.97

However, on a close scrutiny of the Purāṇic regional descriptions one comes to the conclusion that by the Gandhamādana the Purāṇics meant the northern ridge of the great Hindūkush arch with its northern extension, the Khwāja Mohammad range. The Hindūkush, as we know, consists of two parallel ranges which come closer to each other at the apex of the arch, south of the Pāmirs (Meru). These ridges are well-defined in Afghanistan, less so in Kashmir and reappear again as such on the east. The Purāṇics called the northern ridge Gandhamādana and the southern, which is higher and more consistent, Niṣadha, which merged into the Northern Karakoram and Kunlun on the east. The Purāṇic statements that the Gandhamādana is the south, east or west are reconciled if we keep in mind the correct alignment of the northern range of Hindūkush. This range touches the Pāmirs in the south and falls away from it towards southwest and southeast. On the other hand the position of Mālyavān (or Sarikol range) to the east of Meru (Pāmir) is confirmed by statements given in other chapters of the Purāṇas.¹⁰ Variations in its position ((opposite the Gandhamādana) are due to the variations in the position of the Gandhamādana which was not properly interpreted or visualised by subsequent writers.

Linked texts (17)