Pramāṇa · Search · About · API

The Geography of the Puranas c11: Chapter X Bharatavarsa: Janapadas or Communities

The Geography of the Purāṇas c11

Language
English

Contents

Chapter X Bharatavarsa: Janapadas or Communities

c11.1

THE Purāṇic lists of 'janapadas' (or territorial units) and people (or communities) are identical for the simple reason that each janapada is named after the community inhabiting it or vice versa. For instance, the Kuru janapada was the territory occupied by the Kurus, while Nāsikyas was the name given to the people occupying the Nasik region. Thus in all Purāṇic accounts the terms, 'people', and 'janapadas' are synonymous and interchangeable.

c11.2

For the following discussion on the identification and location of the janapadas of Bhāratavarṣa¹ we shall follow the Purāṇic classification and sequence. The Purāṇas give the lists of janapadas or people in each major region mentioned in the previous chapter in a certain order which runs as follows:

c11.3

1. The janapadas or the people of Madhya Deśa or Central Region
2. The janapadas or the people of Udīcya or North-Western Region
3. The janapadas or the people of Aparānta or South-Western Region
4. The janapadas or the people of Prācya or Eastern Region
5. The janapadas or the people of Dakṣiṇapatha or Southern Region
6. The janapadas or the people of Vindhyan Lands
7. The janapadas or the people of Mountain Regions.

I. The Janapadas of Madhyadesa

c11.s1.1

The old name of the Upper Gangetic Basin, 'The United Provinces of Agra and Oudh' was not too ugly for use as a label, for it suggested some of the significant disunities typical of this area, though there is enough fundamental unity for it all to be included in a single region, even if we extend its eastern limits down to the Bihar valley—to include at least Patna and the confluences of the Sone and the Gaṇḍak; and its north-western limits to the valley of the Ghaggar which lies beween the Yamunā and the Sutlej.

c11.s1.2

This was the 'Middle Country' of the Epics, the Mahābhārata and the Rāmāyaṇa, as well as of the Purāṇas. ‘The Great Battle’ was fought on its north-western margins, the land of Kṛṣṇa’s early exploits was its western border, the home of Rāma was in Ayodhyā and the centre of Brāhmaṇism was at Vārāṇasī. The Buddha and his creed were born here, and it was the appropriate centre of the empire of Aśoka, traces of his Pāṭaliputra still surviving in the subsoil of Patna. It was the Hindustān of the Muslim historians and Delhi, Agra, Allahabad, Jaunpur and Lucknow were medieval capitals. It has always been the heart of India, typically Indian and securely Indian. Central India was almost as safe a bulwark as the Himālayas; both the north-western highlands (beyond Panjāb) and the south-eastern ocean were remote, and the lands of approach from both were difficult and unattractive. Its security helped to make it, in a ‘land of sanctuaries’, the richest in shrines; its nodality helped to make it the home of Hindi, the nearest approach amongst the Indian languages—to a *lingua franca*; and it has much fundamental unity—in structure and relief, in climate and its vegetational controls, in the density and occupation of its people.²

c11.s1.3

It is to be remembered that the whole Upper Gangetic Basin, as defined above, was once mostly covered with woodlands and forests. The earliest Āryan settlements were on the rivers amidst cleared or open lands. It was from these nuclei that the territories of various tribes or communities who had settled there grew into compact units and led to the foundation of the respective janapadas and kingdoms. Naturally, within a uniform plain the boundary lines or frontier zones between the different territories were either the broad flood plains of the great rivers or the thick forests which remained uncleared for a long time. Some of these forests (Suyaṃjatavana) remained the normal feature of the land. For instance, the Kuru-jāṅgala was a wild region in the Kuru realm (discussed below), which extended as far as the Kamyaka forest. The kingdom of Uttara-Pañcāla was founded in this jungle tract. The Añjanavana at Sāketa, the Mahāvana at Vaisāli and the Mahāvana at Kapilavāstu were natural forests, which extended up to the slopes of the Himālayas. The Parileyyakavana was an elephant-forest at some distance from Kauśāmbī on the way to Saravasti. The Lumbinivana on the Rohiṇī, the Nāgavana of the Vajji kingdom, the Salavana of the Mallas at Kusinārā, the Bhesakalavana in the Bharga kingdom, the Siṃsapavana at Kauśāmbī, the forest to the north of Setavya in Kośala and the one near Alvi and the Pippalivana of the Moriyas may be cited as typical examples of natural forests which covered a substantial area of the Basin.

c11.s1.4

The Purāṇic janapadas of the ‘Middle Country’ can therefore be defined in terms of two natural features, the rivers and the forests. They are detailed below:

c11.s1.5

*The Janapadas of the Gangetic Doabs*

c11.s1.6

Kuru. The land of the Kurus was originally the hinterland of Delhi west of the Yamunā where they first settled. Later they pushed their boundary towards the east across the Yamunā and the Doab to the Gaṅgā. The Purāṇic janapada of Kuru may be identified with the region bounded by the Ghaggar in the west, the Gaṅgā on the east and the forest belts in the north and south. Their two capitals, Indraprastha on the Yamunā and Hastināpur on the Gaṅgā are well known.

c11.s1.7

Jāṅgala. This janapada is usually associated with the Kurus and is sometimes called Kuru Jāṅgala. It may therefore be inferred that Kuru Jāṅgala was the wasteland of the Kurus and that it occupied the wooded north-eastern part of their territory, which extended up to the Kamyaka forest in the north.

c11.s1.8

Pāñcāla. The janapada of Pāñcāla may be considered as coterminous with Rohilkhaṇḍ of modern times, with the central portion of the Yamunā-Gaṅgā doab added to it. It extended from the Gaṅgā in the west to the Sarju on the east and occupied only the northern half of their doab and the middle portion of the Yamunā and Gaṅgā doabs. It had two divisions, the northern with its capital at Ahicchatra and the southern, with its capital at Kampila. They have been identified as Rāmnagar (in the Bareilly district) and Kāmpilya (in the Furrukhabad district) respectively. The dividing line between the Northern and Southern Pāñcāla was the river Gaṅgā. The southern boundary of the Pāñcāla territory again coincided with the forest belts already mentioned above.

c11.s1.9

Kośala. The Kośala janapada approximately coincided with the Sarju-Rapti doab. It also had two capital cities, Sravasti and Sāketa. According to the Epics and some Buddhist works, Ayodhyā seems to have been the earliest capital and Sāketa the later. In Buddha's time Ayodhyā became an unimportant town, but Sāketa and Sravastī were two of the six great cities of India. Sravastī has been identified with Sahet-Mahet which lies on the boundary of the Gonda and Bahraich districts, 19 miles from Balrampur. At times Kośala extended further south beyond the Sarju and included not only the Gomatī-Gaṅgā-Sarju triangle but also absorbed the janapada of Kāśī.

c11.s1.10

Kāśī. It was one of the ancient janapadas with its capital at Vārāṇasī and was counted among the sixteen Mahājanapadas of India. It extended to and included the southern portion of the Gaṅgā-Gomatī doab and also the Trans-Gaṅgā tract bounded in the north and south by forest belts.

c11.s1.11

*The Janapadas South of the Yamunā-Gaṅgā Axis*

c11.s1.12

Magadha. Further east of the Kāśi-Kosala janapadas beyond the Son and south of the Gaṅgā was the janapada of Magadha. It was bounded by the Son on the west, the Gaṅgā on the north and the spurs of the Vindhyan plateau which touch the Gaṅgā near Monghyr on the east, and Vindhyan forests on the south. This janapada should normally be included in the eastern lands but since it lies on the borders of the 'Eastern' and 'Middle Countries' and also because the territory of Magadha annexed lands to the west of the Son, the Purāṇas thought it fit to include Magadha too in the 'Middle Country'.

c11.s1.13

Kuntala. This janapada has been located by Cunningham to the south of the Gaṅgā near Chunar. If this is correct the Kuntala janapada could correspond to the Mirazapur District of U.P. between the Gaṅgā and Son rivers. A major part of the Gaṅgā-Son doab was covered with forests and woodlands.

c11.s1.14

*The Janapadas of the Trans-Yamunā Region (Western)*

c11.s1.15

Before we attempt the identification of those janapadas which were located in this region, i.e., approximately in the present east Rajasthan and which were included in the 'Middle Country' of the Purāṇas, it would be helpful to take note of its physical framework.

c11.s1.16

The dominating features of Rajasthan are the Aravallis and the desert of Thar. The main Aravalli range runs from Gujarat in the south-west to Delhi in the north-east. Its steep front to the desert plains of Thar is formed of discontinuous and sometimes echeloned ridges, the highest point, the great granitic mass of Mt. Abu (5,646 feet) lies off the main axis in the extreme south-west. Around Udaipur, the Aravallis reach their culmination in a great node of spurs and curving ridges; thence another series of low ridges strike off east-north-east. These ridges run parallel to and south of the Banas river, which cuts across them towards the south, near Tonk, to join the Chambal. They then run further towards the Gambhīra river to the Bharatpur region, from where they turn towards the north parallel to the Yamunā and pass through Deeg to Alwar. In Alwar, the eastern and the main western ridges meet to form a complex of low hilly country from where a spur of the same system meets the Yamunā at Delhi.

c11.s1.17

Another series of hills, low and sometimes detached, belonging to the Vindhyan system runs from Chitor to Fatehpur Sikri (in Agra district) north of the Chambal. The most pronounced ridge of this system is the Bundi Range which effectively cuts off the Kotah basin from the northern regions.

c11.s1.18

Thus the whole of Rajasthan may be divided into the following physical divisions:

c11.s1.19

i. The Alwar-Jaipur region.

c11.s1.20

ii. The Bharatpur-Dholpur-Karauli region.

c11.s1.21

iii. The Bundi-Tonk region.

c11.s1.22

iv. The Mewar or Udaipur region.

c11.s1.23

v. The Trans-Aravalli region occupying the western slopes of the main Aravalli range.

c11.s1.24

vi. The Kotah-Jhalawar basin.

c11.s1.25

In Hiuen Tsang's time No. (i) above was called Vairāta; No. (ii) was part of Sūrasena, later called Braja; No. (iii) and (iv) were known as Vadari and No. (v) as Gūrjara. Nos. (i) to (v) are included by the Purāṇas in the Madhyadeśa and No. (vi) in the Vindhyan lands.

c11.s1.26

*Matsya.* The Matsya janapada was obviously the Vairāta region of Hiuen Tsang's time. It corresponded to physical division (i) above. It was bounded in the north by the broad tangled mass of the Aravalli hills which cover north Alwar and Gurgaon districts, in the east and west by the two ranges of the Aravalli system mentioned above and in the south approximately by the axis of the middle Banas. Its capital, Vairāta (or Bairāta) was located in the Alwar hills at a strategic point where the headwaters of the Sabi and Banganga nearly meet.

c11.s1.27

*Sūrasena.* This janapada coincided with region (ii) above. It was separated from Matsya by the eastern spurs of the Aravallis which run from Lalsot to Deeg in Bharatpur. Its southern boundary ran along the Vindhyan spurs parallel to the left bank of the Chambal. On the base of this funnel-shaped janapada was located their capital Mathurā on the Yamunā. The apex of this funnel pointed the route of Gujarat and western India which for long served as a marchland from the Yamunā basin to the western coast.

c11.s1.28

*Sālva and Associated Janapadas: Bhuliṅga, Bhadrakāra and Bodha*

c11.s1.29

The Sālvas were an important people of ancient India. According to the Mahābhārata, the Sālva country was situated near Kurukṣetra. The Sālvas are nearly always mentioned with Matsyas in ancient literature. Commenting on Salvayayava, the Kasika on Panini's Aṣṭādhyāyī named Udumbara, Tilakhala, Madrakāra (the Bhadrakāra of the Purāṇas), Yugandhara, Bhuliṅga and Śaradanda as the six Avayavas or parts of Śālva janapada. The Vāyu and Matsya Purāṇas locate the Śālvas amongst the people of Madhyadeśa, but the Viṣṇu and Brahma place them in the west. The obvious conclusion is that the Śālva janapada was located somewhere on the boundary of the 'Middle' and 'Western' regions of India.

c11.s1.30

Again, the Śālvas and associated tribes are mentioned with those of the Panjāb, e.g. the Trigartas, the Kekayas, the Madras, etc. Thus the approximate location of the Śālva group of janapadas should be to the west of the Matsya country (which was an integral part of Madhyadeśa), somewhere near the Kuru land and in proximity of the Panjāb janapadas.

c11.s1.31

In addition to this there is another pointer to this group; one of its constituents, the Bhuliṅga has been mentioned by Ptolemy as Bholingai which has been located on the western slopes of the Aravallis.

c11.s1.32

If we refer back to the physical division of Rajasthan mentioned earlier, we will observe that the Śālva group of janapadas pertained to region (v) above. The southernmost janapada, the Bhadrakāra corresponded with the Godwar region west of the Aravallis round about Bhadrajan while the Bhuliṅga corresponded with the upper basin of river Luni which was no doubt a precarious tract on the western slopes of the Aravallis coinciding with eastern Jodhpur.

c11.s1.33

The Śālva (proper) janapada occupied the Sheikhawati-Loharu-Bhiwani region and Bodha corresponded with the Hansi-Hissar-Sirsa tract bounded by the desert on the west, the Śālva on the south and the old bed of the Sutlej in the north. Thus in other words, the Bhuliṅga, Śālva, Bhadrakāra and Bodha janapadas occupied the borderland of the Middle Country so that some of the Purāṇas include them in Middle Country while others count them among the janapadas of the 'Western Lands'. It is to be noted that the precarious tract of Bhuliṅga has been omitted altogether by many Purāṇas.

c11.s1.34

Physical divisions (iii) and (iv) remain unrepresented in the above scheme of Purāṇic janapadas for the obvious reason that they have been included by the Purāṇas in the Vindhyan janapadas detailed below.

II. The North-Western Janapadas

c11.s2.1

A glance at the list of north-western janapadas given in the Purāṇas would at once show that they intend to enumerate the sub-regions of the entire basin of the Indus as well as the regions of the Makran Coast, which are ethnically, politically and geographically associated with that basin.

c11.s2.2

The Indus basin extends from the land drained by the Sutlej in the east to the catchment area of the Kabul river on the west and from the northernmost arc of the Indus to the Arabian Sea Coast. It also includes the mountain rim on its west, down which quite a few streams and rivers flow into the Indus, though some of them lose themselves in the sandy tract which intervenes between the main river and the mountains.

c11.s2.3

A closer examination of the list reveals that the Purāṇic janapadas covered the following zones or sub-regions of the Indus region as defined above:

c11.s2.4

i. The western mountainous zone. This included the present Makran, Baluchistan and Waziristan.

c11.s2.5

ii. The northern and north-eastern mountain (or the Himālayan) zone drained by the affluents of the Kabul and the Indus rivers.

c11.s2.6

iii. The western bank of the Indus from its confluence with the Kabul river to the delta of the Indus.

c11.s2.7

iv. The eastern bank of the Indus from the Panjanād (where all the Punjab rivers meet) to the sea.

c11.s2.8

v. The Punjab plains.

c11.s2.9

vi. The delta.

c11.s2.10

Before we attempt to locate the Purāṇic janapadas in each of the sub-regions of the Indus basin, it would be useful to reiterate the relationship between topography and population incidence which plays an important part in the whole of the basin.

c11.s2.11

From the human point of view, the availability of water for drinking and irrigation is the primary factor in the location of settlements and human groups in regions of low rainfall. The basin of the Indus is one of the regions where rivers big and small control the concentration and distribution of population so forcefully that its population pattern is clearly reflected in its drainage pattern, in fact the two are practically identical. The drainage control is so emphatic that the size of a population group can be translated in terms of the magnitude of the river or stream which served it. From the functional point of view therefore we can divide the rivers into two classes:

c11.s2.12

(1) Small rivers or streams, such as we find in the desert belt or in the mountainous belt to the west of the Indus, are lines of assimilation which attract population all around them. The river valley is therefore invariably associated with a human group which gets settled there, acquires distinct traits and forms a ‘tribe’ or community which in its turn organises itself into a janapada. These janapadas are obviously ‘river janapadas’, separated from each other by spurs of mountains or patches of desert sand.

c11.s2.13

(2) Major, wide and deep rivers with extensive flood plains separate human groups. Here population groups occupy their high banks, settle as near the river as practicable but avoid its floodplain altogether. The river serves as a boundary line between the groups on its both sides and each population group and subsequently each janapada is limited to and defined by the ‘doab’ of two such adjoining rivers.

c11.s2.14

This classification of rivers has a further significance. The tribes or communities living in isolated valleys of small streams or rivers in the northern or western mountain tracts, have always been able to retain their identity and individuality, but while tribes living on the banks of major rivers or in plains between them had better chances to expand, enlarge or consolidate their territories, they were vulnerable to forces from without which led to their expulsion or even extinction. Thus, historically the small cultural or territorial units in mountain valleys have been more ‘stable’ than the bigger units in the plains which may be termed ‘mobile’. It is for this reason that we find discrepancies in the location of the latter in the accounts of the Mahābhārata, Kāvyamīmāṃsa, or even in various Purāṇas as well as in the Greek and the Arab literature. On the other hand, there is a considerable amount of unanimity among the various ancient and medieval works regarding the districts or janapadas occupied by the tribes of the hilly regions in the north and west. We can legitimately doubt if the area indicated in the Mahābhārata as that occupied by the Śūdras or Abhīras is the same to which the Purāṇas refer or if it at all corresponds to the Sodrii or Abiria of Ptolemy, but we can safely assume that the land of the Darudas, Aurasas or Lampākas is the same as Dardrai, Arsa or Lambatai referred to in Greek works.

c11.s2.15

In the light of the above observations we shall analyse the list of north-western janapadas of the Purāṇas and arrange them on a regional basis, i.e., according to the sub-regions of the Indus Basin, indicated in the preceding paragraphs.

c11.s2.16

*The Janapadas of the Western Mountainous Zone*

c11.s2.17

(a) The Makran Region. i. One of the regions of the Makran is the basin of the river Hingol, a winter torrent named Tomeros by the Greeks. Its principal town Hinglaz contained among others an important shrine of Siva which used to draw large crowds of pilgrims from India. The shrine appears to have been in existence even before the days of Alexander.

c11.s2.18

In medieval times when Makran was under the Arabs, Hinglaz rose in importance and was one of the principal towns of those regions. The Purāṇas recognise it as a janapada under the name Aṅgaloka.

c11.s2.19

ii. A region adjoining the Hingol valley is that of the Parikan river. Herodotus calls the inhabitants of this region as the Parikanoi. Parikanoi is a Greek transcript of the Persian form of Parikan, the plural of the Sanskrit Parva-ka or in other words the Ba-rohi or men of the hills. The Purāṇas appear to designate this janapada by the name Pallava (Pahlava or Palnava in some Purāṇas).

c11.s2.20

iii. The valley of the Dasht river, though an insignificant river now, may have supported a comunity or tribe in ancient times. This valley and the adjoining valley of the Bahu river, on whose mouth Gwador is situated, gained considerable importance under the Arabs who considered Makran as the threshold of India and whose main routes from Bampur (south-eastern Persia) to Debel (Indus Delta) traversed these valleys behind the coastal ranges. Ibn-i-Haukal, Istakhri and Idrisi mention the grandeur of their towns, Kiz, Rasak and Faharpara, etc. The Purāṇas name these valleys the janapadas of Deśamaṅka (the valley of the Dasht) and Bahu Bhadra (the valley of the Bahu).

c11.s2.21

iv. The next region in Makran towards the Indus Delta is that of the river Purabi known as Lasbela (the Arabil of the Greeks). The city of Armabel (or Armail or Karabel) is at present only an insignificant town, but picturesquely perched on the banks of the Purali river, it was an important centre during the Arab occupation and commanded for hundreds of years the coastal high-road to India. The whole of the valley is rich in ancient sites like Gondakahar (pre-Muslim) and Gondrani (Buddhist) and was part of a Buddhist kingdom which extended from Armabel to Gandava (near Kalat). The Purāṇic janapada named Hārabhūṣika or Hāramūrtika most probably corresponds to this valley.

c11.s2.22

The mouth of the river Hab and the Churma islands which lie near it formed at one time a stage on the coastal sea-route from the Persian Gulf to the Indus Delta, and were mostly inhabited by seamen or pirates. The Carmakhaṇḍa of the Purāṇas may be identified with this region.

c11.s2.23

(b) Baluchistan and Waziristan. i. Kālatoyaka of the Purāṇas is apparently the Kalāt region of Baluchistan. It practically coincides with the valley of the Malla river which almost touches the foot of the bounding ranges.

c11.s2.24

ii. Vahlika, Vālhika or Bahlika. Balis was a well-known region of Baluchistan, south of Quetta and the Bolan Pass, whose rivers, though not reaching the Indus, drain its basin. It has been described as ‘a district in the desert. It is a place with much cultivation but with few amenities. It contains a few towns, such as Sfinjāi, Kūshk, Sivi of which Kūshk is the residence of the Amir.’ It was also known as Balistan. Vahlika of the Purāṇas evidently corresponds to the region covered by the valleys of the Bolon, Nari and Gokh rivers. It almost coincided with what was formerly known as British Baluchistan, the land of the Balochis.

c11.s2.25

Vahlika has been identified by many scholars with Bactria which is obviously wrong. Bactria belonged to the basin of the Oxus which lies north of the Hindūkush and which according to the Purāṇas, as discussed in earlier chapters, was in the Ketumāla region of the Jambūdvīpa and not in Bhāratavarṣa. It was far beyond the limits of the latter. Besides, whenever the Purāṇas intend to mention regions or people outside the territorial limits of Bhāratavarṣa, they expressly mention this fact to avoid confusion.

c11.s2.26

iii. North of the Vahlika region, we come to the valleys of the Zhob, the Kundar and the Gomal. The upper Gomal has always supported one of the most famous people of the region, the Waziris, who have retained their identity and independence even up to the present time. The Purāṇic Vāṭadhāna or Vāḍhadhāna can possibly be identified with this region.

c11.s2.27

iv. Further north we come across the important valleys of the Kurram and its affluent, the Tochi. The upper Kurram irrigates the Thal region, one of the most beautiful spots in north Waziristan which is probably identical with the Talagāna janapada of the Purāṇas. The Tochi river has given its name to the tribe which occupies its valleys, the Tochis. The Toshara janapada of the Purāṇas most probably meant the land of the Tochis. The identification of Toshara with the Tokharistan region of northern Afghanistan is apparently incongruous when considered in the context of the regional division of the Indus Basin.

c11.s2.28

v. North of the Talagāna and adjoining the Peshawar cul-de-sac is the land of a well-known tribe, the Afridis. It is certainly the same as the Aprytae of Scylax or the Aprita of the Purāṇas and the Mahābhārata.

c11.s2.29

The Janapadas of the Northern Mountain Zone of the Indus Basin

c11.s2.30

(a) The basin of the Kabul river. Practically all the tributaries of the Kabul river which reach it from the northern mountains are associated with tribes or communities which are in most cases synonymous with the Purāṇic janapadas of the same name. Thus we have the following associations:

c11.s2.31

*Gāndhāra* — associated with the lower Kabul valley.

c11.s2.32

*Satadruja* — associated with the valley of the Swat river or the land of Swatis or Assakenians of the Greeks.

c11.s2.33

*Dārva* — The valley of the Pañjkora, which meets the Swat river downstream, has been the home of another frontier tribe with its capital at Dir. It may be identified with the Dārva janapada of the Purāṇas.

c11.s2.34

*Kāmboja* — associated with the valley of the Kunar river. It is to be noted that the Purāṇas do not mention Kāmboja as a part of the Indus basin. They merely say, ‘colonies of Kambojas’. This is true since their land Kāmadeśa or Kafiristan lies across the northern watershed of the Kunar river. However, they might have established colonies in the Kunar basin which have been noted by the Purāṇas.

c11.s2.35

*Lampāka* — associated with the upper Kabul valley. The region is the same as the Lambagae of the Greeks and Lamghan of today.

c11.s2.36

(b) *The Northern and North-Eastern Mountain (Himālayan) Zone.* The following intermontane valley region which are fairly well known have rightly been included by the Purāṇas in their list of the janapadas of the Indus Basin, since all of them lie on the affluents of the Indus which join it on its left or eastern bank.

c11.s2.37

i. Aurasa is the same as ancient Urusa or the modern Hazara district of the North-Western Frontier Province of Pakistan.

c11.s2.38

ii. Darada is still the land of an ancient tribe known as Daradas who live in the valley of river Kiṣengaṅgā in Kashmir. Shardi is its principal town.

c11.s2.39

iii. Kaśmīra or the Kashmir valley drained by the Jhelum.

c11.s2.40

*The Western Bank of the Indus from the Kabul River (Attock) to the Delta of the Indus (from North to South)*

c11.s2.41

i. The lowland belt which runs parallel to the Indus along its western bank has always been an extremely dry and sandy tract. Ptolemy mentions only about a dozen towns of some importance from Amb (Embolina) to Karachi (Kolaka). There were, however, a few communities which occupied the valleys of western streams which could find their way into the Indus in spite of the intervening desert belt. One of such rivers is the Kurram whose valley was the territory of the Vātadhānas already mentioned. In the south of this there appears a town named Paradabathra in Ptolemy's list of stages on the route parallel to the west bank of the Indus. The name is indicative of a janapada mentioned in the Purāṇas as Pārada which from the Greek accounts appears to coincide with the Mithankot region of Dera Ghazi Khan district of the Punjab (Pakistan).

c11.s2.42

ii. South of the Pārada and west of the Indus was the Sindhu janapada which extended along the bank of the river as far south as Sahwan, an important stage on the north-south route.

c11.s2.43

*The Janapadas of the Eastern Bank of the Indus from the Pañjanad to the Delta*

c11.s2.44

i. The Purāṇas mention two janapadas which, according to various references, appear to have occupied the eastern littoral of the Indus. One of them was the Sūdra (Xodraka of the Greeks, expressly mentioned by Ptolemy as a region on the left bank of the Indus). The Sūdras, according to the Mahābhārata occupied the region where the Saraswatī disappeared in the desert. If we take the dry bed of the Hakra as that of the Saraswatī, which is probable, we can identify the Sūdra janapada with the Uch-Khanpur region of Bahawalpur.

c11.s2.45

ii. Further downstream was the land of Sauvīras who from all accounts occupied a territory contiguous to that of the Sindhus but on the opposite bank of the river. Sauvīra, therefore, coincides with the Rohri-Khairpur region of Sind.

c11.s2.46

iii. To the south of Sauvīra, but again east of the Indus where it starts flowing in a number of deltaic streams, was the land of Abhīras. The janapada of Abhīra (Aberia of the Periplus) therefore practically covered the western part of the Hyderabad district of Sind.

c11.s2.47

*The Janapadas of the Punjab Plains*

c11.s2.48

The name Punjab or Pañjanad was applied originally only to the five great rivers east of the Salt range—in the days when the Beas did not join the Sutlej up to Bahawalpur. All of these rivers have always been of great importance, the distribution and movements of the people depended wholly and directly on them, and though their courses have been changing and still change within certain limits and their shoals shift, they carry a considerable amount of water and have always formed, with few exceptions, boundaries between the janapadas of ancient India and also the districts of today. But the boundaries have been largely related to the vagaries of the rivers which have been changing their courses all the time. For instance, in the extreme south of the Sind-Sāgar Doāb, the Indus changed its course over a canal dug from the Chenab to the Indus, and the Chenab had to make a new confluence with it about 60 miles further south. Multan once stood on an island in the 'Ravi,' and Tamerlane's men could work down the Ravi into the Chenab; but the Ravi joins the Chenab now above Multan—fully thirty miles above.

c11.s2.49

Fig. 15

c11.s2.50

BHĀRATAVARṢA JANAPADAS OR COMMUNITIES

c11.s2.51

ABOVE 600 METRES

c11.s2.52

300 - 600

c11.s2.53

BELOW - 300

c11.s2.54

MILES 50 0 50 100 150 200 MILES

c11.s2.55

A characteristic of the doābs of the Punjab is the occurrence of 'high banks' between pairs of rivers. These not only provided excellent sites for towns and cities but also checked to some extent the shifting of the channels. Thus, the Jhelum and the Chenab could never swing west of the Jhelum bank (right bank) or east of the Chenab bank (left bank); and the same is true of the right bank of the old Beas and the left bank of the Sutlej. In fact, the high bank of the Jhelum runs the whole way along its western flank, isolating it wholly from the Indus.

c11.s2.56

Before the irrigation canals came into existence, the population distribution in the doābs was always concentrated in the northern belt of a doāb, i.e., the piedmont zone of the mountains. It is no wonder therefore that in general the capital cities of the ancient janapadas of the Punjab were located on strategic sites at the foot of the mountains.

c11.s2.57

The following is a brief analysis of the Purāṇic list of the janapadas of the Punjab and their relation to the major rivers and their doābs.

c11.s2.58

i. Gāndhāra has already been mentioned along with the Trans-Indus janapadas. In fact, it extended across the upper Indus to the line of hills which radiate from Rawalpindi to the south-west, thus encircling the north-western corner of the Sind-Sāgar Doāb which faced towards the west rather than to the east.

c11.s2.59

ii. Another sub-region of the Sind-Sāgar Doāb is the basin of the river Sohan which is completely enclosed by the Salt Range on the east and the valley of the Indus on the west. The northern part of this basin, i.e. the Rawalpindi and Pindi Gheb region appears to coincide with the Sainika or Piḍika janapada of the Purāṇas, while the southern basin, i.e. the Talagang-Minawali region was probably unpopulated on account of aridity.

c11.s2.60

iii. Adjacent to the above and to its east, was the janapada named Jāguda or Jāṅgala which coincided with the southern half of the Jhelum-Chenab Doāb. 'Jāṅgala' usually indicated an arid bushy country according to the Purāṇas.

c11.s2.61

iv. The northern portion of the Jhelum-Chenab Doāb comprised the famous janapada of Kaikeya whose capital Rajgarha or Girivraja has been identified with Jalalpur by Cunningham.

c11.s2.62

v. The Ravi-Chenab Doab was the land of the Madrakas whose capital Sakela has been identified with Sanglawala Tibba by Cunningham.

III. The South-Western Janapadas

c11.s3.1

The Matsya Purāṇa enumerates the following janapadas which are designated as the ‘Western Janapadas’:

c11.s3.2

(a) (i) Bhārukaccha which has been rightly identified as the Broach region. To be precise, this coastal janapada was located north of the estuary of the Narmadā and extended to the south of the Māhī.

c11.s3.3

(ii) Samahīya janapada is associated with the Māhī river and appears to have been contiguous to Bhārukaccha. The relief feature which demarcated the two janapadas in the dead-level coastal plain could only be the lower Māhī which forms a wide estuary at its mouth. It appears to have included the whole basin of the Sabarmatī situated north of the Māhī.

c11.s3.4

(iii) Saraswata janapada coincided with the region drained by the river Saraswatī which falls into the Rann of Cutch. It corresponds to the Patan-Mehsana plain between the foothills of the Aravallis and the Cutch.

c11.s3.5

(iv) The Saraswata adjoined the Arbuda janapada which was situated to its north-east and occupied the south-western spurs of the Aravallis which contained and enclosed the headwaters of the rivers Sabarmatī and Banas. It may have coincided with the Sirohi-Kotra-Palanpur triangle.

c11.s3.6

(v) Kachika janapada was no doubt the Cutch region.

c11.s3.7

(vi) The Kathiawar peninsula was divided into two janapadas, Ānarta and Saurāṣṭra, the former occupying the northern half and the latter the southern—both separated by the thickly forested mountain core of the peninsula.

c11.s3.8

(b) A few more janapadas have been included by other Purāṇas (except Matsya) in the western region of Bhāratavarṣa, but Matsya locates them in the south and includes them in its list of southern janapadas. The conclusion is that Matsya prefers Narmadā as the natural and cultural boundary between southern India and the rest of the country. Other Purāṇas carry this boundary further south to include the Tāptī valley and, as we will see presently, their boundary followed the crest of the hills (Sahyadri, Gawilgarh, Ajanta, Mahadeo and Maikal) which extend in a general east-west direction south of the Tāptī.

c11.s3.9

The janapadas which have been added by some Purāṇas on account of the shifting of the dividing line between north and south India are as follows:

c11.s3.10

(i) Surāla (Sirāla, Murala). This janapada in all probability coincided with the lower Tāpti basin centred round Surat and Navasari which command this region. This identification has a purely regional basis and needs confirmation from extra-geographical sources.

c11.s3.11

(ii) Tāpasa (Tāmasa, Svāpada) janapada was identical with the Khandeś region which played an important rôle in medieval history. It covered the whole of the middle Tāpti basin between the Sahyadri and Satpura hills and lay athwart the east-west route, along the Tāpti valley and the subsidiary southern route to upper Godāvari. Its core appears to have been roughly the quadrilateral formed by the towns of Savda (near Bhusawal), Pacora, south-west of Jalgaon, Songir and Tapal as shown on modern maps.

c11.s3.12

(iii) Further up the Tāpti valley there appears to have been another janapada, Turiamina which was drained by the upper Tāpti. This janapada is likely to have covered the whole valley of the Tāpti between Badnur and Burhanpur and approximately coincided with southern Nimār.

c11.s3.13

(iv) The valley of the river Purnā, a tributary of the Tāpti, which is separated from the former by the Gawilgarh Hills was divided into two territories. The middle and lower Purnā valley formed the Rūpasa janapada, and the upper Purnā valley, the Kāraskara. The latter occupied the southern foothills of the Gawilgarh hills near the existing Elichpur and Karasgaon towns in northern Amarāvati.

c11.s3.14

(v) Nāsikya appears to be associated with the present town of Nasik and the janapada probably covered the whole of the Darna basin in which Nasik is situated. The Vāyu Purāṇa adds that, besides these janapadas, there are others also which are located on the banks of the Narmadā but their names are not mentioned. This statement shows that the above janapadas are exclusive of the Narmadā janapadas, which is correct since none of the janapadas mentioned above happens to lie in or about the Narmadā valley.

c11.s3.15

(c) The following janapadas have been added to the list of western janapadas by some Purāṇas (Mārkaṇḍeya, Vāyu and Brahmāṇḍa) but have been omitted altogether by the Matsya. This shows that these janapadas were added when more details regarding the western coast of India became available.

c11.s3.16

(i) Sūrparaka (or Sūryāraka) is evidently the western coastal plain drained by the river Sūrya which runs parallel to the coast from north to south and enters the sea near Bassein. It approximately coincides with the Thana district of Maharashtra. In fact both the names of this janapada as given by different Purāṇas are correct. The sites of Safale, Nala Sopara and Sopara which occur within this region are marked on modern maps.

c11.s3.17

(ii) Kālavana (Kolavana) appears to be associated with the town of Kalvan which stands on the Girna river, a southern tributary of the Tāpti. The Kālavana janapada, if so identified, covered the Girna valley up to the point where it descended from the plateau into the plain of the Tāpti (Khāndes), i.e., up to Chalisgaon. It was separated from the Nāsikya janapada by the Chandor range and the Surāla janapada by the crest of the Western Ghats.

c11.s3.18

(iii) There are only two regions in the western coastal plain which remain unaccounted for in the above scheme of identifications. One is the coastal plain between the mouths of the Tāpti and Narmadā rivers and the other between the Surāla and Sūyāraka janapadas. The former may be the Kuliya janapada drained by the Kim river and the latter the Durgā. If this assumption is correct one is tempted to identify the modern Daman-Gangā with the Durgā river of the Purāṇas.

IV. The Eastern Janapadas

c11.s4.1

The ‘Prācya’ or ‘Eastern Country’ of the Purāṇas may be broadly defined as the basins of the lower Gaṅgā and Brahmaputra rivers. The western boundary of this major region may be taken roughly as the course of the river Gaṇḍak from the Himalayan foothills to its confluence with the Gaṅgā; then along the river Son from its mouth, which was formerly near Patna, to the latitude of Rohtas. Thereafter it followed the crest of the Hazaribagh plateau across Bihar and then continued along the south-western boundary of the present State of West Bengal to the sea.

c11.s4.2

The entire ‘Eastern Country’ barring the hilly fringe is essentially a level plain on which innumerable streams wander along ill-defined courses in all directions. There has been a progressive eastward shift of the Gaṅgā water from a main outlet, Bhāgīrathī-Hooghly along the western margins of this alluvial plain to the present main course, the Padmā-Meghnā, with such streams as the Ichamati, Jalangi, Matabhanga, Gorai representing intermediate (not necessarily successive) positions of the most important channel. Thus there is a contrast between the decayed western deltaic alluvium scarred with silted or stagnant bhils, the ‘remains’ of dead rivers and the active eastern deltaic alluvium where the mass of water and silt brought by the rivers is still building the delta. Old traditions and Ptolemy’s observations show that the Bhāgīrathī was the main outlet of the Gaṅgā even as late as the early centuries of the Christian era.

c11.s4.3

The ‘Eastern Country’ as defined above may be conveniently divided into the following natural regions:

c11.s4.4

(a) The Middle Gangetic Valley. This may be further subdivided into two regions:

c11.s4.5

(i) The region north of the Ganges between the Rapti and the Kośī;

c11.s4.6

(ii) The region south of the Gaṅgā, between the Son (in the west and the spurs of the Rājmahāl hills, which touch the Gaṅgā near Rājmahāl town and round which the Gaṅgā turns abruptly towards the south (in the east), the Gaṅgā in the north and the Rājmahāl hills in the south.

c11.s4.7

(b) The Kośī-Gaṅgā and Brahmaputra-Yamunā Doāb watered by the streams flowing down the Himālayas. The usual flat surface of this region is broken near its centre by a stretch of low hills called the Barind, covered with brushwood jungle—the remain of a former forest.

c11.s4.8

(c) The Middle Brahmaputra valley (now forming part of northern Assam).

c11.s4.9

(d) The Gangetic Delta proper between Bhāgīrathī-Hooghli, Gaṅgā-Padmā and the sea.

c11.s4.10

(e) The Western Margin consisting of:

c11.s4.11

(i) The piedmont plain between the Bhāgīrathī-Hooghly axis and the Peninsular block on the west;

c11.s4.12

(ii) The Contai Coastal Plain.

c11.s4.13

(f) The Eastern Margin, i.e., the plain east of the Yamunā-Padmā consisting of:

c11.s4.14

(i) The Yamunā-Meghnā Doāb;

c11.s4.15

(ii) Plains east of the Meghnā.

c11.s4.16

(g) The hilly region between the Rājmahāl and Hazaribagh ranges.

c11.s4.17

The Purāṇic janapadas of the Eastern Country very nearly coincided with the regions mentioned above. This is natural since in the early history of the settlement of this area, population distribution was mainly controlled by natural features such as large rivers, thick forests, inhospitable swamps and hills. Centres of population, which formed the nuclei of the later jana padas were bound to occur at suitable spots in the lowlands where there was the least danger of floods, but where the natural features provided facilities for defence and communication.

c11.s4.18

In the light of the above observations we shall attempt to determine the location and extent of the different Purāṇic janapadas with reference to the natural regions mentioned above.

c11.s4.19

(a) The Janapadas of the Middle Gangetic Valley

c11.s4.20

i. North of the Gaṅgā—Malla, Videha

c11.s4.21

ii. South of the Gaṅgā—Magadha, Aṅga

c11.s4.22

Region (a) (i) north of the Gaṅgā mentioned above, i.e. the Middle Gangetic plain between the Himālayan Terai and the Gaṅgā, the Rapti and the Kośi was occupied by two janapadas. The Malla janapada coincided with the Doāb of the Gaṇḍak and the Rapti-Gaṅgā, while the Videha janapada extended to its east from the Baṇḍak to the Kośi. The capital cities of the Malla janapada were Pawa and Kuśinagar which have been identified by Cunningham with Padrona and Kasia about 35 and 37 miles east of Gorakhpur respectively. The capital city of the Videha janapada was Mithilā about 35 miles north-west of Vaiśālī (modern Basarh in the Muzaffarpur district in Bihar).

c11.s4.23

Region (a) (ii) south of the Gaṅgā contained two janapadas. The janapada between the Son (on the west) and the spurs of the Peninsular Block which press against the Gaṅgā at Mukamah (on the east), the Hazirabagh crest (in the south) and the Gaṅgā (in the north) was known as Magadha. The other, to the east of Mokamah spur and west of the Mandargiri Hills and bounded by the Gaṅgā in the north and the Rajmahal Hills in the south was the Aṅga janapada. The capital of the former was Rajgarha (or Rajgir) situated on a terrace of Barabar Hills while that of Aṅga was Campā (near Monghyr) located within a bend of the Gaṅgā which virtually surrounded the city on three sides.

c11.s4.24

(b) The Kośi-Gaṅgā and Brahmaputra-Yamunā Doāb

c11.s4.25

This triangular region bounded by the Kośi and the Gaṅgā in the west, the Brahmaputra-Yamunā in the east and the Duars on the north is a vast plain diminishing in height from about 300 ft. to 100 ft. above sea-level, scarred by innumerable old river-courses and subject to disastrous floods, especially along the Tistā, which like the Kośi is an exceptionally violent stream. In the heart of this doāb lies the Barind, a large island or terrace of older alluvium. The Barind was formerly covered with thick forest which formed a tongue of the Himalayan forest belt protruding far towards the south. Barind was therefore a huge block of negative tract in the centre of the Doab surrounded by population groups or janapadas between the forest and the bounding rivers.

c11.s4.26

*Pundra.* Pundras are associated with the Angas, Vangas, Kalingas and Magadhas, Suhmas, etc., in ancient literature, e.g. the Mahābhārata, Harivamsa and the Purāṇas. Rājaśekhara (*Kāvyamīmāmsa*) places the Pundra country in the east along with Prāgjyotiṣa and Tamralipta. On the authority of Hiuen Tsang's description of Pundravardhana, Cunningham identified its capital with Mahasthan, 7 miles north of the modern town of Bogra. The Pundra country may therefore be identified with that portion of the Gaṅgā-Brahmaputrā Doāb which was bounded by the two Yamunās on the east and west, the Gaṅgā in the south and the ancient Barind forest in the north.

c11.s4.27

*Malada.* This janapada corresponded to the tract which extended between the Barind forest and the Gaṅgā and included the present Malda district of Bengal as well as the districts of Rajshahi and west Dinajpur of East Pakistan.

c11.s4.28

(c) *The Middle Brahmaputra Valley*

c11.s4.29

This corresponds to the Pragjyotisa janapada of the Purāṇas. This was a famous janapada on the fringe of the 'Eastern Country'. According to *Raghuvamsa* this janapada lay evidently north of the Brahmaputra river and therefore probably included the whole belt of alluvial land north of the Brahmaputra from the Tistā eastwards.

c11.s4.30

(d) *The Delta Proper*

c11.s4.31

The delta proper or the deltaic triangle lies between the Bhāgīrathī-Hooghly (the main outlet of the Gaṅgā in ancient times) and the Padmā. This appears to be coterminous with the Vaṅga country of the Purāṇas. This identification is not only based on regional relationships but is also supported by Ptolemy's account of Gangariddee country and Kālidāsa's *Raghuvamsa* in which he describes Raghu's exploits in these lands.

c11.s4.32

(e) *The Western Margin*

c11.s4.33

The Western Margin, i.e. the territory between the Gaṅgā-Bhāgīrathī-Hooghly axis and the eastern margin of the Bihār plateau.

c11.s4.34

This region was divided into two janapadas with Suhma and the Tāmraliptika. The latter was limited to Midnapur including the Contai Plain District or the Haldi river basin and extended to the sea. North of it there was the Suhma country, mainly covering the districts of Hooghli, Burdwan, Eastern Birbhum and Western Murshidabad. The whole of the Western Margin as defined above comprised the Radha country of later times, which included the Suhma and Brahmottara janapadas of the Puranas. In fact the Matsya uses the name Suhma janapada only in place of Brahmottara, mentioned by Markandeya, Brahma and Vayu. Other Puranas mix the two together and call it the Suhmottara janapada.

c11.s4.35

(f) The Eastern Margin

c11.s4.36

(i) The Yamuna-Meghna Doab appears to have formed the Bhargava-Angaya janapada of the Puranas. It adjoined the Vanga janapada on the west but was separated from it by the Yamuna-Padma.

c11.s4.37

(ii) Regions east of the Meghna are not represented in the Puranic scheme of janapadas. This confirms the view that the eastern limit of the country was the Meghna river.

c11.s4.38

(g) The Hilly Regions

c11.s4.39

Three janapadas appear to have been recognised by the Puranas in the hilly plateau region of Southern Bihar.

c11.s4.40

Mudgarka may have been a janapada coinciding with the north-eastern spurs of the Rajmahal Hills, i.e., approximately the eastern Santal Parganas, southern Monghyr and Bhagalpur districts of Bihar. It appears to be associated with the Mudgagiri hills of Monghyr. Antargiri and Bahirgiri janapadas appear to have been located on the plateau proper. The former occupied the hilly country between the Rajmahal and Hazaribagh ranges and the latter the region beyond the Hazaribagh range, i.e. the basin of the Damodar. Both the janapadas essentially belonged to the Hills.

V. The Southern Janapadas

c11.s5.1

The list of the Puranic janapadas of the South shows that the Daksinapatha or Southern India was normally conceived as that part of the Indian Peninsula which lies south of the crests of Satpura-Mahadeo-Maikal Ranges and the western and southern limits of the Mahanadi Basin. North of this line were the Vindhyan lands while to its north-west lay the western lands of the Puranas. By virtue of their position, the janapadas located on the border lands of these major regions could be assigned to either side. Some Puranas, for instance, the Matsya and the Vayu, rightly consider the Narmada and Mahanadi axis as the dividing line between Northern and Southern India and include the Tapti basin within the southern region.

c11.s5.2

Some included the basin in the western, others in the Vidhyan region. But these assignments do not in general affect the location of the various janapadas with reference to the natural features of the country.

c11.s5.3

Southern India, defined as above, can be divided into the following well-defined regions:

c11.s5.4

(a) The Western Coastal Plain, between the crest of the Western Ghats and the Arabian Sea;

c11.s5.5

(b) The Deccan Plateau;

c11.s5.6

(c) The Eastern Coastal Plain, between the Deccan Plateau and associated hills and the Bay of Bengal.

c11.s5.7

(a) *Janapadas of the Western Coastal Plain—the Western Littoral*

c11.s5.8

*Kerala and Setuka.* South of Kanara the structure and climate of the coastal plain between the scarp of the Western Ghats and the Arabian Sea give it a distinctive personality. The region is known as Malabar—a land of great heat, high humidity, lagoons, sand dunes and greater width of the lowland. It had for centuries the world monopoly of pepper and cardamom. The Malabar coast down to the southernmost tip of the peninsula had always remained an independent territory isolated from the rest of India by virtue of its position between the Ghats and the sea. An Aśoka edict mentions two independent kingdoms of the southern peninsula, viz. that of Keralaputra or the Malabar Coast, and that of Satiyaputra, probably equivalent to the Satyamangalam province of the later kingdom of Madura. Kerala of the Purāṇas, therefore, corresponds to the present Kerala State while Setuka corresponds to its eastern hinterland near the cardamom hills.

c11.s5.9

*Vanavāsaka.* A clan or family called Kadamba enjoyed independent power in the districts now called north and south Kanara and in western Mysore from the third to the sixth century. Their capital Banawāsi, also known as Jayanti or Vaijayanti, was so ancient that it is mentioned in the edicts of Aśoka and in the inscriptions of the Sātavāhana. The Purāṇic janapada Vanavāsaka most probably corresponds to this region. Ptolemy refers to it as Banaunsei located on the western coast.

c11.s5.10

(b) *Janapadas of the Deccan Plateau or the Peninsular Interior*

c11.s5.11

*Māhiṣaka, Kuntala and Kupatha (Kumāra)* janapadas belonged to the region commonly known as Karṇāṭaka, i.e. the real Karṇāṭak or the modern Mysore and not the Madras littoral. This region extends from the Deccan Lavas on the north to the Moyar in the south; the western limit is, of course, the crest of the Ghats and to the south-east the border hills and scarps of the Mysore plateaus provide a fairly sharp boundary between the Moyar and the Palār. But in the north-east, the high plateaus break down into isolated basins south of the Pennar and the Cuddapah ranges. In the north the Raichur Doāb, between the Kṛṣṇa and Tungabhadrā, was the marchland between the Deccani Sultanates and Vijayanagar, which city occupied a forward position on the Tungabhadrā, near Hampi village. This doāb is still a marchland between Karnātaka, Telengāna and to a lesser extent, Mahārāṣṭra.

First 168 of 268 passages. Read on

Linked texts (14)