Pramāṇa · Search · About · API

A Study in the Economic Condition of Ancient India c4: Chapter I. Territorial Divisions in Ancient India

A Study in the Economic Condition of Ancient India c4

Language
English

Contents

Chapter I. Territorial Divisions in Ancient India

I. Regions and Countries

c4.s1.1

THERE are many problems connected with the territorial divisions of ancient India. Among them two are very important, viz. (1) the numbers and dimensions of Indian countries as given by Hiuan Tsang and (2) the numerical designations found in inscriptions, qualifying the names of countries and of territorial divisions, which have been interpreted as meaning the numbers of "villages, towns, and cities" comprised therein. "We are familiar with the fact," says Dr. J. F. Fleet in his article on the Dimensions of Indian Cities and Countries (*JRAS*., vol. xxxix, 1907, p. 641), "that Hiuen Tsiang has usually, if not quite always, indicated the dimensions of the various countries described by him. He has done so by stating sometimes the length and breadth, sometimes the circuit. For instance, he has told us that Kan-t'o-lo, Gandhāra—the capital of which was Pu-lu-sha-pu-lo, Purushapura, Peshawar—measured about 1,000 *li* = 121·21 miles from east to west, and 800 *li* = 96·96 miles from south to north (Julien, *Mémoires*, i, 104). So, again, he has told us that the Cheh-ka, Takka, country—the old capital of which was She-ka-lo, Sākala, Sialkot—had a circuit of about 10,000 *li* = 1212·12 miles (*ibid*., 189). And he has usually indicated in a similar manner the dimensions of the capital cities; thus, he has told us that Pu-lu-sha-pulo and She-ka-lo had circumferences of about 40 *li* and 20 *li* = 4·84 and 2·42 miles respectively. Whence did Hiuen Tsiang obtain these details? Was it from official records, or from the Buddhist books, or from what other source?"

c4.s1.2

He may have obtained the figures either from the official records or from the people on the spot. Other details of a topographical character given in his narrative are found to have been so carefully and accurately recorded that it seems unreasonable to suppose that he should have been careless in respect of the territorial divisions and their dimensions; and we should rather seek to discover whether the correct interpretation of his figures has hitherto been given. If we could ascertain the source from which they were obtained, it might be possible to explain their meaning, and so remove the difficulties which scholars have found in reconciling them with known facts.

c4.s1.3

It appears, so far as my investigations have gone, that, at least from the seventh century to the eleventh century A.D., ancient India was divided into five regions and eighty-four countries (deśa).¹ These five regions consisted of (1) Northern India (Uttarāpatha), (2) Western India (Paścād-deśa), (3) Central India (Madhya-deśa), (4) Eastern India (Pūrva-deśa), and (5) Southern India (Dakṣiṇāpatha).² On account of this five-fold division the Chinese traveller Hiuan Tsang, who travelled through India in the seventh century A.D., called it by the name of the "Five Indies".

c4.s1.4

¹ Caturāśītir deśāḥ. Gauḍa-Kanyakubja-Kaullāka-Kaliṅgāṅga-Vaṅga-Kuraṅgacālya-Kāmākṣa-Oḍra . . . Ābhīra - Narmadātaṭa - dvipa - deśāś ceti (Kāvya-śikṣa, by Vinayacandra, quoted in the Kāvya-mīmāṃsā, by C. D. Dalal in his note on page 24.)

c4.s1.5

² Pṛthūdakāt parata uttarāpathaḥ . . . Devasabhāyāḥ parataḥ pūrvadeśaḥ . . . Tatra Vārāṇasyāḥ parataḥ pūrvadeśaḥ . . . Māhiṣmatyāḥ parato dakṣiṇā-pathaḥ . . . (Kāvya-mīmāṃsā, p. 94.)

c4.s1.6

Northern India comprised the Panjab, Kashmir, and Eastern Afghanistan, and included twenty-one kingdoms, namely (1) Śaka, (2) Kekaya, (3) Vokkāṇa, (4) Hūṇa, (5) Vāṇāyuja, (6) Kāmboja, (7) Vāhlīka, (8) Vahlava, (9) Lampāka, (10) Kulūta, (11) Kira, (12) Taṅgaṇa, (13) Tuṣāra, (14) Turuṣka, (15) Barbara, (16) Harahūva, (17) Hūhuka, (18) Sahuḍa, (19) Hamsa-mārga, (20) Ramaṭha, and (21) Karakaṇṭha. Perhaps the Sarasvatī river may have been the boundary between Northern and Central India.

c4.s1.7

Western India comprised Western Rajputana, Sindh, Kacch and Gujarat, and contained ten kingdoms, namely (1) Deva-sabhā, (2) Surāṣṭra, (3) Daśeraka, (4) Travaṇa, (5) Bhṛgukaccha, (6) Kacchīya, (7) Ānarta, (8) Arbuda, (9) Brāhmaṇavāha, and (10) Yavana. The Narmadā river on the south, the Devasabhā on the east, the Sarasvatī river on the north and the sea-coast on the west were the boundaries of Western India.

c4.s1.8

Eastern India comprised the modern Bihar, Bengal, Assam, and Orissa, and the ancient Kaliṅga-rāṣṭra. It included sixteen kingdoms, namely (1) Aṅga, (2) Kaliṅga, (3) Kosala, (4) Tosala, (5) Utkala, (6) Magadha, (7) Mudgara, (8) Videha, (9) Nepāla, (10) Puṇḍra, (11) Prāgjyotiṣa, (12) Tāmaliptaka, (13) Malada, (14) Mallavartaka, (15) Suhma, and (16) Brahmottara.

c4.s1.9

Southern India comprised the whole of the country from Kaliṅga to Koṅkaṇa and from the Narmadā to Ceylon, and included twenty-one kingdoms, namely (1) Mahārāṣṭra, (2) Māhiṣaka, (3) Aśmaka, (4) Vidarbha, (5) Kuntala, (6) Kratha-kaiśika, (7) Śūrpāraka, (8) Kāñcī, (9) Kerala, (10) Kāvera, (11) Murala, (12) Vānavāsaka, (13) Siṃhala, (14) Coḍa, (15) Daṇḍaka, (16) Pāṇḍya, (17) Pallava, (18) Gāṅga, (19) Nāsikya, (20) Koṅkaṇa, and (21) Kollagiri.¹

c4.s1.10

¹ Pṛthūdakāt parata uttarāpathaḥ. Yatra Śaka-Kekaya-Vokkāṇa-Hūṇa-Vāṇayuja-Kāmboja-Vālhika-Valhava-Lampāka-Kulūta-Kira-Taṅgaṇa-Tuṣāra-Turuṣka-Barbara-Harahūva-Hūhuka-Sahuḍa-Hamsamārga-Ramaṭha-Kara-kaṇṭha-prabhṛtayo janapadāḥ Himālaya-Kulindrendra-Kila-Candrādayaḥ parvatāḥ . . . . Devasabhāyaḥ parataḥ paścāddeśaḥ. Tatra Devasabha-Surāṣṭra-Daśeraka-Travaṇa-Bhṛgukaccha-Kacchīyānartārbuda-Brāhmaṇavāha-Yavana-prabhṛtayo janapadāḥ . . . Tatra Vārāṇasyāḥ parataḥ pūrvadeśaḥ. Yatrāṅga-Kaliṅga-Kosala-Tosalotkala-Magadha-Mudgara-Videha-Nepāla-Puṇḍra-Prāgjyotiṣa-Tāmaliptaka-Malada-Mallavartaka-Suhma-Brahmottara-prabhṛtayo janapadāḥ . . . Māhiṣmatyāḥ parato dakṣiṇāpathaḥ. Yatra Mahārāṣṭra-Māhiṣakāśmaka-Vidarbha-Kuntala-Krathakaiśika-Śūrpāraka-Kāñcī-Kerala-Kāvera-Murala-Vānavāsaka-Siṃhala-Coḍa-Daṇḍaka-Pāṇḍya-Pallava-Gāṅga-Nāsikya-Kauṅkaṇa-Kollagiri-Vallara-prabhṛtayo janapadāḥ . . . (Kāvya-mīmāṃsā, pp. 93-4.)

c4.s1.11

Central India comprised the eastern portion of Rajputana and the whole of the Gangetic plain from Thāneśar to Benares.

c4.s1.12

Rājaśekhara, the poet, does not give the names of its component kingdoms, but there is no doubt that, with the help of Hiuan Tsang, these names could be restored. It probably consisted of the following sixteen kingdoms:—(1) Kurukṣetra, (2) Matsya, (3) Śrughna, (4) Maḍavara and Brahmapura, (5) Goviṣāṇa, (6) Ahicchatra, (7) Pilośanā and Sām-kāśya, (8) Śūrasena, (9) Kānyakubja, (10) Ayodhyā, (11) Hayamukha, (12) Prayāga, (13) Kauśāmbī, (14) Kuśapura, (15) Vaiśākha, and (16) Kāśī. Not having access to the original Saṃskṛta sources, Sir Alexander Cunningham included some portions of Eastern and Western India with Central India : in reality it contained only sixteen kingdoms, and not thirty-seven, as he thought.

c4.s1.13

As to the geographical division of Ancient India into nine regions, as given by Varāhamihira and in the Purāṇas, it was perhaps made from the astronomical point of view, and adopted by poets and literary folk for providing similes with the lotus, and sometimes with the human body, which contains, according to the Hindu belief, nine ‘entrances’ for the senses. The Śrīcakra-śambara-tantra may be quoted as a good instance of how these nine divisions were used to furnish similes.¹ The Bodhāyana-gṛhya-sūtra went so far as to specify the ruling planet together with the name of the country representing each main division.²

c4.s1.14

¹ Śrīcakraśambara Tantra, pp. 30-1.

c4.s1.15

² Atha nava-graha-pūjā-vidhiḥ. Madhye vartulākāra-maṇḍale pratyaṅmukham Kaliṅga-deśajam. . . . Sūryasya dakṣiṇa-dig-bhāge trikoṇākāra-maṇḍale dakṣiṇābhimukham Avantīdeśajam. . . . Sūryasya pūrva-dig-bhāge pañca-koṇākāra-maṇḍale prāṅmukham Bhojakaṭaka-deśajam. . . . Sūryasyāgneya-digbhāge caturaśrākāra-maṇḍale pratyaṅmukham Yamunādeśajam Ātreya-gotrajam. . . . Sūryasyaiśāna-dig-bhāge bāṇākāra-maṇḍale prāṅmukham Magadha-deśajam Ātreya-gotrajam. . . . Sūryasyottara-dig-bhāge dīrgha-caturasra-maṇḍala udaṅmukham Sindhu-deśajam Aṅgirasa-gotrajam. . . . Sūryasya paścimadig-bhāge dhanur-ākāra-maṇḍale udaṅmukham Kāśyapa-gotrajam . . . Saurāṣṭra-deśajam. . . . Sūryasya nai-ṛtya-dig-bhāge Sūrpākāra-maṇḍale dakṣiṇābhimukham Barbara-deśajam. . . . Sūryasya vāyavya-dig-bhāge dhvajākāra-maṇḍale dakṣiṇābhimukham Antarvedi-deśajam. . . . (Bodhāyana-gṛhyasūtra, adhyāya 17, praśna 1, pp. 196-205.)

c4.s1.16

In connexion with the number of countries Cunningham separated Persia and Ceylon from India proper, and reduced the number of eighty-two countries, as given by Hiuan Tsang, to eighty; but, as a matter of fact, two kingdoms should be added to make up the number eighty-four, as described by Vinayacandra in his *Kāvya-śikṣā*. Ceylon was for a long time considered to be a portion of India and at one time more definitely as the fourth district of Malaya.¹ It was represented at the *Rāja-sūya* sacrifice of Yudhiṣṭhira, and played an important part in epic poetry. It embraced the Buddhist religion in the time of Aśoka, and always took a leading part in the sea-borne trade of southern India.

c4.s1.17

¹ Tatra Vindhyādayaḥ pratītasvarūpa Malaya viśeṣās tu catvāraḥ. Tesu prathamaḥ :—" Āmūlayaṣṭeḥ phaṇiveṣṭitānām saccandanānām jananandanānām Kakkola-kailā-maricair-yutānām jātitarūṇām ca sa janma-bhūmiḥ." . . . Caturthaḥ :—" Sā tatra cāmikara-ratna-citraiḥ prāsādamālā-valabhir vitaṅkaiḥ. Dvārārgalā-baddha-sureśvarāṅkā Laṅketi yā Rāvaṇa-rājadhānī." (*Kāvya-mīmāṃsā*, p. 92.)

c4.s1.18

Persia was a vast country, and some of its frontier districts were included in the Aparānta, Barbara, and Yavana kingdoms of India. It exported fine horses and other valuable articles to India, and Indian goods to the kingdoms of the west. "They (the people of Palmyra), being merchants," says Appian, "bring from Persia to Arabia Indian commodities, which they dispose of to the Romans."² Kālidāsa included it among the countries conquered by Raghu,³ and the Mauryan Emperor Candragupta annexed a portion thereof to his own empire. "The Indians," writes Strabo, "occupy (in part) some of the countries situated along the Indus, which formerly belonged to the Persians. Alexander deprived the Ariani of them, and established there settlements of his own. But Seleukus Nikator gave them to Sandrokottus, in consequence of a marriage contract, and received in return 500 elephants."⁴ As to its boundary, Pliny says that "most writers do not fix the Indus as the western boundary (of India), but add to it the four satrapies of the Gedrosi, Arachotæ, Arii, and Paropamisadæ—thus making the River Cophes its extreme boundary ”.¹

c4.s1.19

² McCrindle's *Anc. Ind.*, p. 214.

c4.s1.20

³ Pārasīkāṃs tato jetum pratasthe sthala-vartmanā. (*Raghuvamśa*, Canto iv, verse 60.)

c4.s1.21

⁴ Cunningham's *Anc. Geo.*, Ed. 1924, p. 18. Cp. McCrindle, *op. cit.*, pp. 88-9.

c4.s1.22

¹ Cunningham, Anc. Geog. of India, Ed. 1924, pp. 17–18.

c4.s1.23

The division of India into eighty-four countries seems very old. The Bārhaspatya-artha-śāstra gives the number of the principal countries as eighteen, that of the minor countries as seventeen, that of the marshy land countries as twenty-five, that of the hill countries as twenty, and that of the miscellaneous countries as four, making the total number eighty-four. It informs us, further, regarding the area of each country, which in some places is the same as that given by Hiuan Tsang and in other places slightly more or less. The names of the countries, together with the areas, as given by the Bārhaspatya-artha-śāstra,² may be arranged in a table in the following manner:—

c4.s1.24

² 86. Rāmasṛṣṭiś catvāriṁśacchatam dakṣiṇottare āsahyam dvādaśa Viśvāmitrasṛṣṭir ekādaśa. 87. Nepālam catuḥśatam. 88. Pūrva-samudra-tīre varuṇataḥ samudrāntam aṣṭa-yojanā. 89. Pañca-śata-dvitayam Uttaralāṭam Pūrvalāṭam ca. 90. Kāśi-Pāñcāla-dvitayam aśītiḥ. 91. Kekaya-Sṛñjayam ṣaṣṭiḥ. 92. Mātsya-Māgadham śatam. 93. Mālava-Śakuntam aśītiḥ. 94. Kosalāvantiḥ ṣaṣṭiḥ. 95. Saihya-Vaidarbha-dvitayam śata-dvayam. 96. Vaideha-Kauravam śatam. 97. Kāmboja-Daśārṇam aśītiḥ. 98. Ete mahā-viṣayāḥ. 99. Ete khalu caturāśrāḥ. 100. Āraṭṭa-Bālhīkau dakṣiṇottarataḥ śata-mātrau pūrva-paścād dvādaśau. 101. Śaka-Saurāṣṭrau caturasrau catvāriṁśat. 102. Aṅga-Vaṅga-Kaliṅgāḥ śatamātrāś caturaśrāś ca. 103. Kāśmīra-Hūṇāmbaṣṭha-Sindhavah śatamātrāś caturāśrāś ca. 104. Kirāta-Sauvīra-Cola-Pāṇḍyā uttare dakṣiṇe sthitiḥ śatātparaṁ ṣaṣṭi-mātrāḥ. 105. Yādava-Kāñcī-viṣayam catvāriṁśacchata-mātram. 106. Ete upaviṣayāḥ. 107. Sapta-koṅkaṇāś catuḥśatamātrā dvādaśa ṣaḍrāṣṭrau ca. 108. Ete anūpāḥ. 109. Sahyādrau catvāro giriviṣayāḥ. 110. Śrīparvate dvayam. 111. Raivataka ekaḥ. 112. Vindhye pañca. 113. Kumāra ekam. 114. Mahendre trayam. 115. Pārīyātre trayam. 116. Sarve dakṣiṇottarataḥ pañcāśan-mātrāḥ pūrvataḥ paścāt pañca-yojanaḥ samāḥ. 117. Mlecche yavana-viṣayāḥ pārvatīyāḥ. (Le Muséon, 33, pp. 1915–16.)
70. Tatrottare Himavān. 71. Tasya dakṣiṇe Nava-sāhasrī bhuḥ. 72. Tatra dakṣiṇātyo Bhārataḥ khaṇḍaḥ. (Le Muséon, The Bārhaspatya-sūtra, iii, p. 143.)

c4.s1.25

| Countries. | Area in yojanas ? | Countries. | Area in yojanas ? |
| 1. Uttara-Lāṭa | 105 | 1. Āraṭṭa | 100 × 12 |
| 2. Pūrva-Lāṭa | 105 | 2. Vālhika | 100 × 12 |
| 3. Kāśī | 80 | 3. Śaka | 40 |
| 4. Pāñcāla | 80 | 4. Surāṣṭra | 40 |
| 5. Kekaya | 60 | 5. Aṅga | 100 |
| 6. Sṛñjaya | 60 | 6. Vaṅga | 100 |
| 7. Matsya | 100 | 7. Kaliṅga | 100 |
| 8. Magadha | 100 | 8. Kāśmīra | 100 |
| 9. Mālava | 90 | 9. Hūṇa | 100 |
| 10. Śakunta | 90 | 10. Ambaṣṭha | 100 |
| 11. Kosala | 60 | 11. Sindhu | 100 |
| 12. Avanti | 60 | 12. Kirāta | 100-60 ? |
| 13. Saihya | 100 ? | 13. Sauvīra | 100-60 ? |
| 14. Vaidarbha | 100 ? | 14. Cola | 100-60 ? |
| 15. Vaideha | 100 | 15. Pāṇḍya | 100-60 ? |
| 16. Kuru | 100 | 16. Yādava | 140 |
| 17. Kāmboja | 80 | 17. Kāñcī | 140 |
| 18. Daśārṇa | 80 | | |
| Total = 18 Principal countries (mahāviṣaya). | | Total = 17 Minor countries (upaviṣaya). | |

c4.s1.26

| 7 kingdoms of Koṅkaṇa | 104 |
| 12 ,, dvādaśa-rāṣṭra | 104 ? |
| 6 ,, ṣaḍ-rāṣṭra | — ? |
| Total = 25 kingdoms of low lands (anūpa-viṣaya). | |

c4.s1.27

| 4 kingdoms on the Saihya Mt. | 50 × 5 |
| 2 ,, ,, Śrī Mt. | 50 × 5 |
| 2 ,, ,, Raivataka | 50 × 5 |
| 5 ,, ,, Vindhya Mt. | 50 × 5 |
| 1 ,, ,, Kumāra Mt. | 50 × 5 |
| 3 ,, ,, Mahendra Mt.. | 50 × 5 |
| 3 ,, ,, Pāriyātra Mt. | 50 × 5 |
| Total = 20 kingdoms on hills, each having an average area of about 70 yojanas. | |

c4.s1.28

| 1. Rāma-sṛṣṭi | 140, 112 ? |
| 2. Viśvāmitra-sṛṣṭi | 111 |
| 3. Nepāl | 104 |
| 4. Varuṇa | 108 (8 ?) |
| Total = 4 miscellaneous kingdoms. | |

c4.s1.29

Total of countries mentioned in the Bārhaspatya-artha-śāstra = 84 (18 + 17 + 4 + 25 + 20 = 84).

c4.s1.30

There is probably much truth in the figures given by Hiuan Tsang, if we understand them to refer to the area and not to the circumference of each kingdom. Take, for instance, the kingdoms of Gurjara, Valabhi, Surāṣṭra, Ānandapura, and Aṭāli. These, according to Hiuan Tsang, had a circuit of about 26,000 *li*. The *Bārhaspatya-artha-śāstra* gives in respect of these kingdoms certain figures which total 630 *yojanas*, or 25,200 *li* (taking one *yojana* as equal to 40 *li*). The difference of 800 *li* is comparatively small, if we take into account the fact that the boundaries of the kingdoms may have varied between the two periods of time, and the difficulty of identifying exactly the area referred to in each case. It is remarkable that both these authorities give the very same figure in respect of Aṅga, viz. 4,000 *li*, or 100 *yojanas*. In the case of other kingdoms the figures given are exactly the same, provided we take a *yojana* as equal to 12.12 miles or 50 *li*, as suggested by Dr. J. F. Fleet in his article on "The Yojana and Li" (*JRAS*., vol. xxxviii, 1906, p. 1011). It appears that two distinct *yojanas* were in use in Ancient India, namely, one equal to 9.532312 miles, or say 40 *li*, as suggested by a French authority, and the other equal to 12.12 miles, or say 50 *li*, as proposed by Dr. J. F. Fleet. The similarity between the figures of the *Si-yu-ki* and the *Bārhaspatya-artha-śāstra* can be seen from the following table:—

c4.s1.31

The *Bārhaspatya-artha-śāstra* figures and the *Si-yu-ki* figures compared.

c4.s1.32

| Countries. | The figures given in the *Bārhaspatya-artha-śāstra* yojanas. | The figures given in the *Si-yu-ki* li. | |
| Aṅga | 100 (× 40 =) | 4,000 | 1 yojana = 40 li. |
| Vārāṇasī | 80 (× 50 =) | 4,000 | 1 ,, = 50 ,, |
| Kaliṅga | 100 (× 50 =) | 5,000 | 1 ,, = 50 ,, |
| Magadha | 100 (× 50 =) | 5,000 | 1 ,, = 50 ,, |
| Kāñcī (Draviḍa) | 140 (× 50 =) | 6,000 | 1 ,, = 50 ,, |
| Nepāla | 104 (× 40 =) | 4,000 (approximately) | 1 ,, = 40 ,, |
| Cola | 60 (× 40 =) | 2,400 | 1 ,, = 40 ,, |
| Vaṅga (Puṇḍravardhana) | 100 (× 40 =) | 4,000 | 1 ,, = 40 ,, |

c4.s1.33

| Surāṣṭra countries. | | Surāṣṭra countries. | |
| Yādava | . 140 | Aṭālī | . 6,000 |
| Sauvīra | . 140 | Kaccha | . 3,000 |
| Sindhu | . 100 | Valabhī | . 6,000 |
| Valabhī | . 105 | Ānandapura | 2,000 |
| Pūrva-lāṭa | . 105 | Surāṣṭra | . 4,000 |
| Surāṣṭra | . 40 | Gurjara | . 5,000 |
| **Total** | **= 630** *yojanas*. | **Total** | **= 26,000** *li*. |

c4.s1.34

Surāṣṭra countries total = 630 (× 40 = 25,200 *li*) = 26,000 (approximately, taking 1 *yojana* = 40 *li*).

c4.s1.35

The similarity between the figures given in the *Bārhaspatya-artha-śāstra* and the *Si-yu-ki* may possibly afford an answer to the question raised by Dr. Fleet, namely, “Whence did Hiuen Tsiang obtain these details? Was it from official records, or from the Buddhist books or from what other source?” As the *Bārhaspatya-artha-śāstra*, like the *Artha-śāstra* of Kauṭalya, is a treatise dealing with administrative affairs, it is most probable that both authorities derived their figures from the official records kept in each kingdom for the purpose of collecting land revenue. And if this was the source, as it is reasonable to conjecture, then the figures given by both should be taken as representing only the area surveyed by government officers, and as but approximately correct. As to Dr. Fleet’s further question (in his article on “Dimensions of Indian Cities and Countries” quoted above): “And to what extent, in respect of the countries in particular may we, making allowance for the fact that the statements in this class would obviously give only approximations, in round numbers, accept these details as authentic?” If we understand Hiuan Tsang’s figures to represent, not the circuit, but the surveyed area of each kingdom, then their accuracy or inaccuracy can easily be roughly tested.

c4.s1.36

The total for all India of the figures given in the *Bārhaspatya-artha-śāstra* is 9,000 *yojanas*.¹ The same figure is given by other authorities also. The average of the figures given in respect of each kingdom by Hiuan Tsang amounts to approximately 4,000 *li*, or say 100 *yojanas*. If we multiply this by 84, the total number of kingdoms as given in the *Bārhaspatya-artha-śāstra*, we get 8,400 *yojanas*. Or, if we take the figures given in the *Bārhaspatya-artha-śāstra* in respect of thirty-three principal and minor countries (as it does not give round numbers for Vāhlīka and Āraṭṭa), the total is 3,250 *yojanas*, giving an average per country of 98.7 *yojanas*. Multiplying this average by eighty-four, we should get 8,290.8 *yojanas* for the whole of India.

c4.s1.37

¹ Nava-sahasra-yojana-vistīrṇe bhārata-khaṇḍe. Nava-sahasra-yojana-vistīrṇa-bhāratavarṣe. (*Śrī-Śukla-yajurvedīyānām brahma-nitya-karma*, pp. 10, 12.)

c4.s1.38

Sir Alexander Cunningham took a *yojana* to be equal to 9 miles. The *yojana* not representing the same distance in all parts of the country, it will be safe to adopt as the standard *yojana* the average of the different *yojanas*. According to Cunningham, there were three *krośas* prevalent in his time, viz. that of 1¼ miles in the north-west and the Panjab, that of 2¼ miles in the Gangetic provinces, and that of 4 miles in Bundelkhand, Mysore, and Southern India. The average of these three is 2½ miles.¹ As one *yojana* was always recognized to be equal to 4 *krośas*, we may fairly assume the average length of the *yojana* to have been about 10 miles. Therefore 8,628 *yojanas* would represent about 862,800 square miles (taking the figures as referring to area, and not circuit).

c4.s1.39

¹ Cunningham's *Anc. Geo.*, Ed. 1924, p. 658.

c4.s1.40

Modern India, excluding Burma, but including the Native States, comprises about 1,600,000 square miles,² nearly double the area as calculated from the figures given in the *Bārhaspatya-artha-śāstra* and the *Si-yu-ki*. The difference may be explained by the fact that the balance in ancient times was unexplored or unsurveyed, owing to its being covered by forests or uncultivable.

c4.s1.41

² *Census of India*, 1921, vol. i, p. 58; *Statistical Abstract for British India*, 1912–13 to 1921–2; *East India (Statistical abstract)*, vol. 57, 1925, p. 2.

c4.s1.42

Now it seems clear that Hiuan Tsang's figures were based either upon the information given him by the people he consulted on the spot, or upon the records kept by the local authorities. In reducing the *yojana* to *li* he appears to have treated the figures as linear measurements, multiplying by 40, instead of by (40 × 40) 1,600, as he should have done if they represented superficial measure. That Hiuan Tsang, with his great learning and accuracy of observation and record, should have made an error of this kind may perhaps be regarded as the chief stumbling-block to the acceptance of the conclusion that the figures represent areas in square measure.

c4.s1.43

Such a mistake is, however, very common with the ancient writers. For instance, the *Dāna-mayūkha* and many other books treating of similar subjects, we find the *nivartana* measure described as (3 × 10) 30 *daṇḍas*,¹ instead of (3 × 10 × 10) 300 *daṇḍas* as given by Kauṭalya.²

c4.s1.44

¹ Chap. III, p. 83, note 1.

c4.s1.45

² Daśa-daṇḍorajjuḥ. Dvi-rajjukaḥ parideśaḥ. Tri-rajjukaṃ nivartanam. Ekato dvi-daṇḍādhiko bāhuḥ. (Kau. Artha, p. 107.)

c4.s1.46

Saṃskṛta words expressing superficial measure are generally made by adding the prefix *pari*, which means ‘surrounded by’, e.g. *parimāṇa*, *pari-kṣetra*, etc. It is quite likely that the Chinese traveller tried to translate the Saṃskṛta prefix *pari* by the Chinese word *chau*, which had exactly the same meaning, namely ‘surrounded by’.

c4.s1.47

Whatever may be the origin of the error, it is quite clear that in both authorities, the *Bārhaspatya-artha-śāstra* and the *Si-yu-ki*, the figures cannot be taken as representing circuit. But, on the other hand, if we understand them to denote the surveyed areas, we do not meet with any serious difficulty.

II. Smaller Territorial Divisions: Their Names and Meanings

c4.s2.1

The problem regarding the numerical designations attached to the countries mentioned in the last section is more difficult to solve than the problem connected with their number and dimensions. Take, for instance, the Southern Indian inscriptions referring to land grants. They contain numerical designations in connexion with some countries, e.g. the Kisukāḍ seventy (*EI.*, vol. xii, No. 32, p. 296, note by Dr. J. F. Fleet), the Torugare sixty (ibid., vol. xv, No. 6, p. 79), the Māsavāḍi hundred and forty (ibid.), the Sāgar three hundred (ibid., vol. xii, No. 32, pp. 272-3), the Karatikallu three hundred in the Eḍadore two thousand (ibid., No. 34, p. 313), the Noḷambāḍi thirty-two thousand (ibid., vol. xvi, No. 7, p. 28), the Kaṅgāl five hundred (ibid., vol. xvi, No. 7, p. 28), the Hagaritage three hundred (ibid., vol. xii, No. 34, p. 307), the Kukkanūr thirty (ibid., vol. xvi, No. 8, p. 37), the Bāḷakund three hundred (ibid.), the Elambi twenty (ibid., vol. xvi, No. 10, p. 72), the Kandur and Santalige thousand each (ibid., No. 9 B., p. 59), the Banavāsa twelve thousand (ibid.), the Belvolā or Beluvāḷ (ibid., No. 9 A, p. 56), and Huḷigere three thousand each (ibid., No. 8, p. 35), the Palāsige twelve thousand (ibid., p. 50), the Kuṇḍi three thousand (ibid., No. 1, p. 3), the Pānuṅgall five hundred (ibid., vol. xii, p. 298), the Niryumbola seventy (ibid., No. 32, p. 290), the Kundbura thirty (ibid., p. 298), etc. The grand total of these figures in respect of Southern India amounts to seven *lakhs* fifty thousand (see *Bomb. Gaz.*, vol. i, pt. ii, p. 341, n. 2.).

c4.s2.2

What do these numerical components mean? Why are they given so much importance in connexion with land grants?

c4.s2.3

What purpose do they serve there? Are they mythical and exaggerated? "There has been a mistaken idea," says Dr. Fleet, "which apparently originated with Dr. Burnell (see his *South Indian Palæography*, second edition, p. 67, last paragraph but one), that the numerical components of this and similar appellations denote the amount of revenue. And some apparent reason for it might be found in the facts that there are not so many as twenty thousand villages in Mysore, and not quite forty-four thousand villages and hamlets in the whole of the Bombay Presidency. . . But there are quite enough passages to show clearly that the reference is to the numbers, real, exaggerated, or traditional, of the cities, towns, and villages : for instance, the Aihole inscription of A.D. 634-5 mentions ‘the three Mahārāṣṭras, containing ninety-nine thousand villages’ (*Ind. Ant.*, vol. viii, p. 244); the Śilāhāra records of A.D. 1026 and 1095 distinctly speak of a division of the Koṅkaṇ containing ‘fourteen hundred villages’ (id., vol. v, p. 280, and vol. ix, p. 38); an inscription at Pāṭna in Khāndēsh, of about A.D. 1222, speaks as distinctly of ‘the country of the sixteen hundred villages’ (*Epigraphia Indica*, vol. i, p. 345); and the meaning of the name of the territorial division, called the Vēḷugrāma or Vēṇugrāma seventy, is explained by a passage which describes Vēḷugrāma as ‘resplendent with seventy villages’ (*Jour. Bo. Br. R. As. Soc.*, vol. x, p. 252). Other instances of very large numbers are, the Noḷambavāḍi thirty-two thousand, in the district of Bellāry; the Kavaḍidvīpa lakh-and-a-quarter, which was the northern part of the Koṅkaṇ; and the seven-and-a-half lakh country, which is the expression that was used in later times to denote the territory that was held first by the Rāṣṭrakūṭas and then by the western Chālukyas. These large numbers must be gross exaggerations, based possibly on some traditions or myths. But there appears no reason for objecting to accept the literal meaning of such more reasonable appellations as the ‘Koṅkaṇa fourteen-hundred and nine-hundred’, the ‘Sāntaḷige thousand’, the ‘Tardavāḍi thousand’, the ‘Pānumgal five-hundred’, and the ‘Beḷvola three-hundred’; and possibly, when we know more as to how far the larger numbers include the smaller, of the ‘Kūṇḍi three-thousand’, the ‘Karahāṭa four-thousand’, the ‘Toṛagale six-thousand’, the ‘Palasige twelve-thousand’, and the ‘Banavāsi twelve-thousand’. The system of administration by dividing the country into circles of tens, twenties, hundreds, and thousands of villages, is prescribed in the *Mānava-dharma-śāstra*, vii, 113 to 117”. (*Bomb. Gaz.*, vol. i, pt. ii, p. 298, n. 2.)

The Term Grāma

c4.s2.t1.1

From the passage quoted above it appears that Dr. Fleet had no hesitation in accepting the numbers as true when they were small; but, when very large, he thought that they were mythical or exaggerated. Personally I should rather prefer to consider them to be true or false as a whole, whether they be large or small. However, before rejecting or accepting these figures, it will be as well to find out the actual meaning of the word *grāma*, which they qualify. It is translated in English by the words ‘village’, ‘town’, or ‘city’. Does it really signify this or something quite different?

c4.s2.t1.2

When cataloguing the *Prajñāpanopāṅga*, an old Jain canonical work, I was astonished to find the meaning of *grāma* given as an estate or a survey village which can pay eighteen kinds of Government taxes.¹ I consulted many Saṃskṛta lexicons as to this interpretation, but without any success. It is well known that these lexicons were compiled from a literary view-point and that they are not of much assistance as far as the technical meanings of words are concerned. But the case was quite different with the Jain lexicographers, who had to deal with the Prākṛta words used in different meanings in their scriptures.

c4.s2.t1.3

¹ . . . Mahāvidehesu cakkavaṭṭi-khaṃdhāvāresu vāsudeva-khaṃdhāvāresu baladeva-khaṃdhāvāresu maṃḍaliya-khaṃdhāvāresu mahā-maṃḍaliya-khaṃdhāvāresu gāma-ṇivesesu ṇagara-ṇivesesu ṇigama-ṇivesesu . . . (Commentary: Mahāvidehesu, kintu cakravarttiskaṃdhāvāresu . . . “gāma-ṇivesesu” ityādi, grasati buddhyādīn guṇān iti grāmaḥ, yadi vā gamyaḥ śāstra-prasiddhānām aṣṭādaśakaraṇām iti grāmaḥ. . . . Śrī-Prajñāpanopāṅga, fol. 46, part i.)

c4.s2.t1.4

In the time of the great Guptas, and even many centuries prior to them, it appears that the word *grāma* was used in official records for an estate, and in poetical and literary works for a village or settlement. A discussion arose among scholars with regard to its technical meaning. They agreed that the word *grāma* meant an estate, whether cultivated, fallow, or *jangal*, but they differed as to the size or area to which it referred.

c4.s2.t1.5

Like the commentator of the *Prajñāpanopāṅga*, Vijayacandra also in his *Abhidhāna-rājendra*¹ defines *grāma* as an

c4.s2.t1.6

¹ *Gāma-grāma-pum*. Gamyo gamanīyo ’ṣṭādaśānām śāstre prasiddhānām karaṇām iti vyutpattyā, grasate vā buddhyādīn guṇān iti vyutpattyā vā pṛṣodarāditvān niruktavidhinā grāmaḥ. . . Prācuryena grāma-dharmopetatvāt karādigamyo vā grāmaḥ. Ācā. 2 Śru. 1 A. 2 U. . . Sanniveśa-viśeṣe, Praśna. 3 Āsra. dvāra. Jñā. Kaṇṭaka-vāṭakā-vṛtam janānām nivāse, Utta. 2 A. Sūtra. . . Tatrāneka-vidha-naigamānām anyāny api pakṣāṇi vaktavyāni tāni nāma-grāham saṃgṛhṇann āha :—

c4.s2.t1.7

Gāvo taṇāī sīmā, ārāmam udapāṇa-ceṭa-rūpāṇi |
Vāhiya vāṇamamtara, nuggaha tatto ya āhipati.

c4.s2.t1.8

Gāvaḥ 1 tṛṇāni 2 upalakṣaṇatvāt tṛṇāhārakādayaḥ sīmā 3 ārāma 4 udapānam kūpaḥ 5 ceṭarūpāṇi 6 vāhirvṛtiḥ 7 vāṇam antaram devakulam 8 avagrahaḥ 9 tataś cādhipatiḥ 10 iti niyukti gāthākṣarārthaḥ. Atha bhāvārtha ucyate—prathamam naigamaḥ prāha—yāvantam bhūbhāgam gāvaś caritum vrajati tāvan sarvo’pi grāma iti vyapadeśam labhate. Tato viśuddhanaigamaḥ pratibhaṇati—

c4.s2.t1.9

Gāvo vayaṃti dūram, pijam tu taṇakaṭṭhahāragādīyā |
Sūruṭṭhite gataem-ti attha samte tato gāmo.

c4.s2.t1.10

Paristhūram api paragrāmam api caritum vrajanti, tataḥ kim evam so’py eka eva grāmo bhavatu? api ca evam bruvato bhavato bhūyāsām api parasparam atidavīyasām grāmaṇām eka-grāmataiva prasajjati, na caitad-upapannam, tasmān naitāvān grāmaḥ kimtu yat yāvanmātram kṣetram tṛṇāhārakāṣṭhahārakādayaḥ sūrya utthite tṛṇādyartham gatāḥ santaḥ sūrye ’stamayati tṛṇādi-bhārakam baddhvā punar āyānti, etāvat kṣetram grāmaḥ.

c4.s2.t1.11

Parasīmam pi vayaṃti hu, suddhataro bhaṇati jā sa sīmā tu |
Ujjāṇa avattā vā, ukkilaṃ tā u suddhaparo.

c4.s2.t1.12

Śuddhataro naigamo bhaṇati—yadyapi gavām gocara-kṣetrād āsannataram bhūbhāgam tṛṇa-kāṣṭhā-hārakā vrajanti, tathāpi te kadācit parasīmānam api vrajanti, tasmān naitāvān grāma upapadyate. Aham bravīmi—yāvat svā ātmiyā sīmā etāvān grāmaḥ. Tato’pi viśuddhataraḥ prāha—maivam. Atipracuram kṣetram grāma iti vocaḥ, kimtu yāvat tasyaiva grāmasya sambandhī kūpaḥ tāvad grāma iti. Tato’pi viśuddhataro brūte—udyānam ārāmas tāvad grāma iti bhaṇyate. Viśuddhatamaḥ pratibhaṇati—etadapi bhūyastaram kṣetram na grāma-samjñām labdhum arhati, aham bhaṇāmi—yāvad udapānam tasyaiva grāmasya sambandhī kūpaḥ tāvad grāma iti. Tato’pi viśuddhataro brūte—idam apy atiprabhūtam kṣetram, ato yāvat kṣetram avyaktāni ceṭarūpāṇi ramāṇāni gacchanti tāvad grāmaḥ. Tato’pi viśuddhataraḥ prativakti etad apy atiriktatayā na samīcīnam ābhāti, tato yāvantam bhūbhāgam atilaghīyāmso bālakā utkrīḍanto riṅgantaḥ prayānti tāvan grāma iti.

c4.s2.t1.13

Eva viśuddha-nigamassa vai parikkhevaparivumo gāmo |
Vavahārassa vi evam, saṃgaha jahi gāma samavā ō.

c4.s2.t1.14

Evam vicitrābhiprāyāṇām pūrvanaigamānām sarva api prati-pattīrvapoṣya sarva-viśuddha-naigama-nayasya yāvān vṛti-parikṣepaparivṛto bhūbhāgas tāvān grāma ucyate. . . (*The Abhidhāna-rājendra* by Vijayacandra; the article “Grāma”, pp. 865-7.)

c4.s2.t1.15

estate which can pay eighteen kinds of government taxes or which is assessed separately for revenue purposes. The word was used in the remotest period of Indian history in ten different meanings, viz. : (1) cows (gāvaḥ) ; (2) grasses (tṛṇāni) ; (3) boundary (sīmā) ; (4) pleasure-garden (ārāma) ; (5) well (udapāna) ; (6) servants ? (ceṭa ?) ; (7) fences (bahiḥ) ; (8) temple (devakula ?) ; (9) an estate (avagraha) ; (10) owner (adhipati).

c4.s2.t1.16

Take, for instance, the first meaning, ‘cows.’ There was a body of scholars (naigamaḥ) who believed that grāma meant an estate containing an area as large as that of a pasture-ground. They defined grāma as an estate (kṣetra) where cows go to graze. This definition was rejected by others on the ground that cows sometimes entered a grāma (estate) belonging to others, which had its own separate existence as an estate (grāma). Others tried to explain the exact area of a grāma by saying that all land which is traversed by a wood-cutter should be taken as one grāma (estate). This definition again was not accepted owing to its being very vague. Some proposed to take grāma as meaning the estate owned by a family, and others as an estate equal in size to a garden ; and so on. Apart from its technical meanings, there existed no difference of opinion as far as its popular use was concerned, namely, for ‘a group of people settled on a portion of land and vice versa’.

c4.s2.t1.17

The Inscription No. 20 of the Pāṇḍu-Lena Caves of Nāsik furnishes clear proof that the word grāma was frequently used for a petty estate. Understanding the term to mean ‘a village’, the editor of the text was perplexed by the unusual shortness of the inscription, and remarked that “The words in the original are Dhambhika-gāma, and seem to mean the village Dhambhika. As the text stands, this must be accepted, but it is unusual for the people of a city to bestow a village in gift. Villages generally are granted by kings, not by the people. Assuming that the people of Nasik did grant the village, it is curious that the inscription should be so short and that it should make no mention of the person to whom it was given or of the object of grant”. (See Bomb. Gaz., vol. xvi, p. 590, Pāṇḍu-Lena Caves, No. xviii, Inscr. No. 20.) The difficulty here mentioned vanishes if we understand the word grāma as meaning an estate. According to the Dāna-mayukha a donor should at least give a grāma to a Brāhmaṇa; if unable to give so much, he should at least give land measuring a go-carma. From this it is clear that the grant of a grāma to a Brāhmaṇa could be made by anyone. However in such cases the permission of the king was necessary. The ground for this will be discussed in the chapter dealing with the sale and purchase of land.

c4.s2.t1.18

There is another most important inscription in the same place containing the words tac ca kṣetram vikṛṣyate sa ca grāmo na vasati evaṃ sati, meaning “it was thought that that field would be cultivated; but as that field (sa ca grāmo) is not under cultivation so”, clearly showing not only that a kṣetra could be described as a grāma, but also that the word vasati was used in the sense of ‘cultivation’, as well as in that of ‘habitation’.

c4.s2.t1.19

The definition given by the Kāmikāgama, Kāraṇāgama, and Mayamata of the ekabhoga-grāma or kuṭika-grāma also shows that grāma means an ‘estate’. According to them an ekabhoga-grāma means a property or an estate which can be enjoyed by one family, consisting of the owner and his servants.¹

c4.s2.t1.20

¹ Viprair athānyair varṇair vā bhogyo grāma udāhṛtaḥ. Eko grāmaṇiko yatra sabhṛtya paricārakaḥ. Kuṭikantadvijāniyād ekabhogas sa eva tu. Kāmikāgama. Prāg-ukta-guṇa-yuktāya vedārthanipuṇāya ca. Kuryāt taj jñānanidhaye dānam syād uttamottamam. Ekabhogam iti khyātam praśastam atidurlabham. Kāraṇāgama. Anyad aśaktānām ced dānam daśabhūsurāntamekādi. Ekakuṭumbisametam kuṭikam syā(t ta)d ekabhogam iti kathitam. Mayamata. (Quoted in EI., vol. xv, No. 5, p. 55.)

c4.s2.t1.21

It is to be regretted that all the translators of the ancient Indian law books should have committed the mistake of translating grāma everywhere by the English word ‘village’. Owing to this error, as will be pointed out, some very eminent scholars were misled into building up theories regarding the nationalization of land, the organization of village communities, etc.

c4.s2.t1.22

Take, for instance, the Smṛti of Yājñavalkya. According to that work, the disputes about a boundary of a field should be settled by the sāmantas possessing equal grāmas (sama-grāmaḥ = having equal estates).¹ As to the meaning of sāmanta, Kātyāyana says that it means only an owner who possesses a neighbouring estate, whether such be a house, a field or a portion of land (grāma). He says further that if the neighbouring sāmantas are personally interested in any dispute, then it should be settled by the sāmantas next to them, and if they also are not disinterested, then it should be decided by the sāmantas next adjoining. This whole group is termed lotus-shape. It appears from the Artha-śāstra of Kauṭalya that it comprised forty people (sāmanta-catvārim-śat-kulyā) in number, who had the privilege of being present at the time of sale and purchase of land and of settling disputes in connexion with fields, houses and estates.²

c4.s2.t1.23

¹ Sāmantā vā samagrāmāś catvāro’ṣṭau daśāpi vā | Raktasragvasanāḥ sīmām nayeyuḥ kṣitidhāriṇaḥ. (Commentary.) Grāmo grāmasya sāmantaḥ kṣetram kṣetrasya kīrtitam | Gṛham gṛhasya nirdiṣṭam samantāt parirabhya hi. (Kātyāyana.) Saṃsaktakās tu sāmantas tat saṃsaktās tathottarāḥ | Saṃsakta-sakta-saṃsaktāḥ padmākārāḥ prakīrtitāḥ. (Ibid.) (Yājñavalkya-smṛti; Vyavahāra-prakaraṇa 9, śloka 152, pp. 232-3.) Sākṣyabhāve tu catvāro grāmāḥ sīmāntavāsinaḥ | Sīmāvinirṇayam kuryuḥ prayatā rāja-samnidhau. (Manu-smṛti, viii, 258.)

c4.s2.t1.24

² Sāmanta-pratyayā vāstu-vivādāḥ. Gṛham kṣetram ārāmas setu-bandhas taṭākam ādhāro vā vāstuḥ. (p. 166.) Jñāti-sāmanta-dhanikāḥ krameṇa bhūmi-parigrahān kretubhyā bhaveyuḥ. Tato’nye bāhyas sāmantacatvāriṃśatkulyā gṛha-prati mukhe veśma śrāvayeyuḥ. Sāmanta-grāma-vṛddheṣu kṣetram ārāmam setubandham taṭākam ādhāram vā maryādāsu yathāsetu-bhogam anenārgheṇa kaḥ kretā iti trirātra-ghuṣita-vītam avyāhatam kretākretum labheta. (p. 168.) (Kauṭalya’s Artha-śāstra.)

c4.s2.t1.25

If we take *grāma* to mean an estate, no difficulty arises. Everything seems reasonable and natural. All disputes were to be settled by the *sāmantas* of the adjoining estates. According to my interpretation, if a house in a street was to be sold, it was sold in the presence of forty householders residing in the same street or in the immediate vicinity. If a field in a village was to be sold, it was sold in the presence of the landowners (*sāmantas*), forty in number, holding estates in the vicinity of that field.

c4.s2.t1.26

But things take a very different complexion to those who look on *grāma* as signifying a ‘village’. The whole of the above-mentioned text then becomes meaningless. Interpreting *sama-grāmāḥ* as denoting the people possessing equal villages, would mean that disputes, sales, and purchases, etc., connected with a field in a village, were to be settled by the people living in other villages or, in other words, that each village was governed by a group of forty villages. The constituent villages of the group would vary in the case of different villages; and if two or more disputes or sales, etc., were going on simultaneously, the inconvenience and difficulties arising from such a regulation may readily be conceived!

c4.s2.t1.27

Such an obviously absurd system of administration could never have existed in Ancient India. Why should the *sāmantas* of certain villages interfere with the affairs of others, possibly at some distance. Why should they be required to waste their time in travelling such distances in order to give an opinion in a dispute connected with a field or a house wherein they had no interest, and about which they knew nothing whatever? It would have been a great injustice to the land-owning classes of a village to have internal affairs settled by strangers and to count for nothing in their own dwelling-place. Furthermore, it is ridiculous to suppose that landowners could have been forced to wander about from one village to another and neglect their own estates. It is possibly from such considerations as the above that certain scholars have translated *sāmantāḥ vā samagrāmāḥ* as “the neighbours living in the same village”. The use of *sama*, however, in the sense of ‘same’ is unusual, if not doubtful, in Saṃskṛta. The word *sama*, when joined to another word generally means ‘having equal’. ‘Same,’ on the other hand, is ordinarily expressed by the Saṃskṛta word *samāna*, or *tat*.

c4.s2.t1.28

It is interesting also to note how Kauṭalya employed the term. In *prakaraṇa* 173, which deals with the *Paragrāmīkayoga*, he uses it for an ‘estate’; *grāma-ghāta* for laying waste the countryside; *grāma-vadha* or *grāma-avas-kanda* for capturing a fort representing, or counted in the revenue records as, an estate; *para-grāma* for an estate belonging to others, etc. In *prakaraṇa* 171 also, the word *grāma* is used in the same meaning (estate). Groups of five and ten estates, which were under the supervision of a *gopa*, are referred to by the words *pañca-grāmī* and *daśa-grāmī*.¹

c4.s2.t1.29

¹ Svāmino dūṣya-grāmam (p. 401); Grāma-vadhe’vaskande ca (p. 402); Grāma-ghāta-praviṣṭām vā (p. 403); Grāma-ghāta-daṇḍasya (p. 402); Vijigīṣuḥ paragrāmam avāptukāmaḥ (p. 394). See also Prakaraṇas 173 and 171; pp. 394–403. (Kauṭalya’s Artha-śāstra.) Tatpradiṣṭaḥ pañca-grāmīṃ daśa-grāmīṃ vā gopaś cintayet. (Ibid., p. 142.)

c4.s2.t1.30

Going further back to the period of the Vedas and Upaniṣads we find that the word *grāma* was used more clearly for an estate. For instance, in the *Bodhāyana-dharma-sūtra* we find the word *grāma* explained by the commentator as *vāstu*,² which is defined by Kauṭalya as an estate containing land, garden, house and well, etc.³ In the *Chāndogya-upaniṣad*, in the story of Raikva the Brahmajñānin, a deserted place (vijanadeśa) is called a grāma.¹ Saṃgrāmāḥ samitayaḥ of the Atharvaveda perhaps means an assembly of the people possessing estates.² It appears that the word araṇya was used for those estates which were under forest, as it occurs in contrast to the word grāma in several places.³

c4.s2.t1.31

² Nirgatya grāmānte grāma-sīmānte vā’vatiṣṭhate tatra kuṭīṃ maṭham vā karoti kṛtam vā praviśati. (Commentary.) Grāmānto vāstu-sīmā. Itaraḥ kṣetra-sīmā. Kuṭī eka-sthūṇam veśma. Matho bahu-sthūṇaḥ. . . (Bodhāyana-dharma-sūtra, iii, i, sūtra 13, p. 300.)

c4.s2.t1.32

³ Sāmanta-pratyayā vāstu-vivādāḥ. Gṛham kṣetram ārāma-setu-bandhas taṭākam ādhāro vā vāstuḥ. (Kauṭalya’s Artha-śāstra, p. 166.)

c4.s2.t1.33

¹ So'dhastācchakaṭasya pāmānam kaṣamānam upopaviveśa. . . . (The Śaṃkara-bhāṣya :-ity uktaḥ kṣattā anviṣya tam vijane deśe adhastāc chakaṭasya gantryāḥ pāmānam kharjūm kaṣamānam kaṇḍūyamānam dṛṣṭvā. . . ) tam hābhyuvāda Raikvedam sahasram gavām ayam niṣko'yam aśvatarīratha iyam jāyā yam grāmo yasminnāsse mā bhagavaḥ śadhītī. (The Śaṃkara-bhāṣya :-ayam ca grāmaḥ yasminnāsse tiṣṭhasi sa ca tvadarthe mayā kalpitaḥ. . . ) (Chāndogya-upaniṣad; Adhāya IV, khaṇḍa 1-2, mantras 8 and 4.)

c4.s2.t1.34

² Ye saṃgrāmāḥ samitayas teṣu cāru vadāma te. (Atharva-veda, xii, i, 56.) Saṃgrāme samyatte samayakāmaḥ. (Taittirīya-saṃhitā, ii, i, 8, 4.) For an estate-owner the word grāmaṇī appears to be used in Vedic literature. See :- Grāmaṇyo gṛhān paretya mārutam sapta-kapālam puroḍāśam nirvapati. . . (Śatapatha-brāhmaṇa, 3, 1, 6.) Mārutaḥ saptakapālo vaiśyasya grāmaṇyo gṛhe. (Maitreya-saṃhitā, ii, 6, 5; iv, 3, 8.) Sindhukūlāśritā ye ca grāmaṇiyā mahābalāḥ. Mahā-bhārata, Sabhā-parva, xxxii.)

c4.s2.t1.35

³ Yadgrāme yadaranye yatsabhāyām yadindriye. (Atharva-veda, v, xx, 17.) See also the Chāndogya-upaniṣad, v-x, 3.

c4.s2.t1.36

To sum up, we shall not be far from the truth if we understand grāma to mean an estate, comprising cultivated and fallow lands as well as pasture, and including any buildings, wells, etc., existing thereon; the area, in fact, that constituted the territorial unit for the purpose of revenue assessment.

III. Smaller Territorial Divisions: Their Number and Size

c4.s3.1

From the foregoing meanings of the Saṃskṛta word grāma it is clear that it was used to express many other things besides an inhabited hamlet or a village. In official revenue records it seems to mean an estate which paid government taxes and the king's dues. The "Kundavura thirty" of the copperplate grant can be explained as meaning a place for the collection of revenue which comprised thirty estates.¹ The Miraj plates of Jayasimha II record that the king granted an estate (translated as a village in *Ep. Ind.*, vol. xii, No. 34), called Māḍadūjhūru, belonging to the group of the Karaṭikallu three hundred, which was in the Eḍadore two thousand.² Dr. J. F. Fleet thought that the Eḍadore two-thousand was a stretch of country between the Rivers Kṛṣṇā on the north and Tuṅgabhadrā on the south, comprising a large part of the present Raichur district.³ At present this district contains altogether 893 villages.⁴ Taking the word *grāma* as meaning an estate, we have no reason to distrust the above-mentioned figures. The Karaṭikallu was probably a fairly large village, approximating to a town; and if we divide 2,000 by 300, the number of such towns would come to about six. There were small centres of revenue collection (*saṃgrahaṇa*) like Elambe, Kukkanur, Kisukād, Kundavura, Toṛugare, etc., comprising somewhere between twenty and seventy estates.⁵ The number of such centres of revenue collection comes to nearly forty, if fifty may be taken as the average number of estates in each centre (*saṃgrahaṇa*).

c4.s3.2

¹ *EI.*, vol. xvi, No. 8, p. 35; vol. xii, No. 32, p. 290.

c4.s3.3

² *EI.*, vol. xii, No. 32, p. 296, notes by Dr. Fleet; *IA.*, vol. viii, p. 18; *EI.*, vol. xii, No. 34, p. 303.

c4.s3.4

³ *IA.*, vol. viii, p. 18; *EI.*, vol. xii, No. 32, p. 296, notes by Dr. Fleet; *EI.*, vol. xii, No. 34, p. 303.

c4.s3.5

⁴ *Imp. Gaz.*, vol. xxi, p. 39.

c4.s3.6

⁵ *EI.*, vol. xv, No. 6, p. 79; vol. xii, p. 298; vol. xvi, No. 8 B, p. 37; vol. xvi, p. 28, No. 7; *IA.*, vol. xxx, 1901, p. 259 ff.

c4.s3.7

It has already been shown that we need find no difficulty in explaining the numerical appellations occuring in the Southern Indian inscriptions, provided we take *grāma* as meaning an estate. But it is interesting to note that these same numerical appellations seem to have proved a stumbling-block to writers from 326 B.C. down to modern times. The confusion dates from the time of Alexander the Great, when information about the kingdom of Poros was gathered in detail. “This country, it is said,” writes the Greek historian Plutarch, “contains 15 tribes, 5,000 considerable cities, and villages without number.”¹ General Bunbury formed the same opinion as Dr. Fleet about these large numbers, and could not help remarking that “when the Greek writers tell us that the district between the Hydaspes and the Hyphasis alone contained 5,000 cities (?), none of which was less than that of Cos (Strabo, xv, p. 686), and that the dominion of Poros, which was confined between the Hydaspes and the Acesines—a tract not more than 40 miles in width—contained 300 cities (ibid., p. 698), it is evident that the Greeks were misled by the exaggerated reports so common with all Orientals, and which were greedily swallowed by historians of Alexander with a view of magnifying the exploits of the great conqueror”.²

c4.s3.8

¹ McCrindle, *The Invasion of India by Alexander the Great*, p. 309.

c4.s3.9

² Ibid., p. 309, note; Bunbury, *Hist. of Anc. Geog.*, i, p. 453.

c4.s3.10

Taking the *grāma* as synonymous in meaning with the English word ‘estate’, we can readily understand the system of distributing land followed for revenue purposes by the Hindu kings. It appears that in ancient times, a country (*deśa*) was divided into *janapadas*, *janapadas* into *gaṇas* or say *parganas*, and *gaṇas* into *grāmas*. For instance, the *rāṣṭra* of Vanavāsaka contained 12,000 estates (*grāma*); that of Nolambavāḍi (Nulumbavāḍi), 32,000; that of Gangavāḍi, 96,000; that of the three Mahārāṣṭras, 99,000; and that of Southern India, 750,000.³ Vinayacandra in his *Kāvya-śikṣā* gives the numbers of estates for Eastern, Western, and Central India, as follows: (1) Koṅkaṇa, 1,414; (2) Candrāvatī, 1,800; (3) the *janapada* on the banks of the Mahī river, 2,200; (4) Surāṣṭra, 9,000; (5) Lāṭa-deśa, 21,000; (6) Gurjara-deśa (including Aparānta), 70,000; (7) Ahūḍa and Brāhmaṇa-pāṭaka (both together), 100,000; (8) Dāhala, 900,000; (9) Mālava-deśa, 1,892,000; and (10) Kānyakubja (empire ?), 2,600,000. After giving these figures, he adds that the countries the numbers of which are not given should be taken as belonging to northern and southern India.¹ For this reason I take the number 2,600,000 as representing grāmas (estates) of the Kānyakubja empire. It is interesting to note that the countries mentioned above are very much the same as those which constituted the empire of Harṣavardhana.

c4.s3.11

³ Bomb. Gaz., vol. i, pt. ii, p. 298, n. 2.

c4.s3.12

¹ Kauṅkaṇa-prabhṛti-caturdaśādhikāni caturdaśa-śatāni. Candrāvatī-prabhṛti aṣṭādaśa-śatāni. Dvāviṁśati-śatāni mahitaṭam. Nava-sahasrāṇi surāṣṭrāḥ. Ekaviṁśati-sahasrāṇi lāṭa-deśaḥ. Saptati-sahasrāṇi gurjara-deśaḥ pārataś ca. Ahūḍa lakṣāṇi brāhmaṇapāṭakam. Navalakṣāṇi ḍahalāḥ. Aṣṭādaśa-lakṣāṇi dvinavatyadhikāni mālavo eśdaḥ. Ṣaḍ-viṁśal-lakṣāṇi kānyakubjaḥ. Anantaram uttarāpatham dakṣiṇāpatham ceti. (Kāvya-śīkṣā by Vinayacandra, quoted by C. D. Dalal in his introduction to the Kāvya-mīmāṁsā, p. 24, note.)

c4.s3.13

As to the authenticity of these figures, they can be verified from other sources. Take, for instance, the Koṅkaṇa 1,414 of Vinayacandra. The Bhāṇḍūp plates of Chittarājadeva mentions a town named Purī, “which is marked as the chief town of a province consisting of fourteen hundred villages” (grāmas or estates).² The Thāṇā plates of A.D. 1017, the Bhādāna grant of A.D. 997 and the Khārēpāṭaṇ plates of A.D. 1095 all mention the number of estates (grāmas) in Koṅkaṇa as 1400.³ After careful study, Dr. Fleet arrived at the conclusion that “the Purī or Koṅkaṇa fourteen hundred seems to answer fairly closely to the Bassein, Sālsette, Bhiwṇḍi, and Kalyāṇ tālukas, with perhaps also the Karjat tāluka of Thāṇā, and the Panwēl, Pēṇ, and Alibāg tālukas of Kolāba”.⁴ For Gujarāt (Gurjara-dēśa) and Mālwā (Mālava-dēśa), Vinayacandra gives the figures as 70,000 and 189,200 respectively. ’Abdullah Wassāf informs us that “it is related by sufficient informants, experienced travellers . . . that Sawālik contains 125,000 cities and villages; and Mālwā 1,893,000 towns and villages. . . Gūjarāt, which is commonly called Kambāyat, contains 70,000 villages and towns, all populous, and the people abound in wealth and luxuries".¹ In the same way we are told by Rashīdu’d-dīn in his book entitled *Jāmi‘u’t-tawārīkh* that "It is said that Guzerat [Gujarāt] comprises 80,000 [? 70,000] flourishing cities, villages, and hamlets. . . After these comes the country of Sawālak, which comprises 125,000 cities and villages. After that comes Mālwāla, which means 1,893,000 [? 1,892,000] in number".² It is possible that the copyist of the *Jāmi‘u’t-tawārīkh* transcribed a 2 in the case of Mālwā and a 7 in the case of Gujarāt as 3 and 8 respectively. Though Vinayacandra does not give any figures in respect of Sawālik, we know from the *Prabhāvaka-carita* of Hemacandra that it contained 125,000 *grāmas*, if we may assume Arṇorāja to have been the king of Sawālik.³ Further, we are told that the country of Jālandhara contained the same number of *grāmas* (estates).⁴ These numerical territorial designations survived down to the Muhammadan period. After conquering Mālwā, says ’Abdullah Wassāf, Sultān ’Alāu’d-dīn despatched Malik Nabu [and] Zafar Khān to conquer the province of Telingāna. When they arrived on its frontier, "the Rāi . . . agreed to pay an annual tribute and receive the royal collectors, and that populous territory . . . containing more than 30,000 tracts of country, was added to the Muhammadan empire."⁵ The expression "tracts of country" used in this passage is remarkable, and seems to support the view that we have here an equivalent of the Saṃskṛta *grāma*.

First 99 of 133 passages. Read on

Cites (40)