Skip to main content Accessibility help

The Beginnings of Civilization in South India

  • Clarence Maloney


The diffusion of Indian Civilization and its “great tradition” to the extreme south of the peninsula occurred in the earliest stages by sea, not by land. Such characteristics of civilization as script and literacy, kingship and the state, and organized religions, developed at first in the earliest urban centers in Tamil Nadu, which were located on the coast opposite Ceylon. In the half millennium before Christ there was sea traffic between the coasts of Gujarat and Sind, and Ceylon, which laid the basis for the development of civilization in that island. Early civilization in soutliern Tamil Nadu developed parallel with this, and most of its intrusive features were analogous with those in the island, or were inspired by them. The earliest attractions of the far southern coasts were pearls and gems, which brought merchants, and ultimately, script, religions, and the Pandiyan dynastic traditon. Early civilizations characterized by the Colas and Keralas, as well as in the Andhra deltas, were inspired by sea traffic. Sources are early Tamil and Ceylon Brahmi inscriptions, Tamil Sangam literature, early Ceylon chronicles, etymologies, legends, and some archeology.



Hide All

1 The wider context, and also more data on this subject are discussed by the author in The Effect of Early Coastal Sea Traffic on the Development of Civilization in South India (dissertation, University of Pennsylvania, 1968).

2 Arthaśāstra ii.11, trans. Kangle, R. P., The Kauṭilīya Arthaśāstra (Bombay, 19631965).

3 Arrian, , Indika 8.

4 For discussion of the date of this part of the Arthaśāstra, see Kangle, , III, 7475, 8687, 106–15. Pearls are referred to in the Brāhmaṇas, though the source is not mentioned, and Atharva Veda iv.10 is a hymn about the healing properties of pearls and mother-of-pearl.

5 Duyvendak, J. J. L., China's Discovery of Africa (London, 1949), pp. 910. The place the Chinese merchants came to is probably identifiable as Kāñcipuram, doubtless served by the easily defended port of Mahabalipuram.

6 Rea, Alexander, South Indian Buddhist Antiquities (Archaeological Survey of India, New Imperial Series, XV, 1894), pp. 711.

7 Periplus Maris Erythraei 59; Ptolemy, , Treatise on Geography, vii.1.10–11; 1.13.1; Pliny, , Natural History, vi.17.23; Aelian, , On the Characteristics of Animals, xv. 9.

8 Pliny vi.17.23; Aelian xv.8.

9 Puranāṉūṟu 380.1; Neḍunalvāḍai 125. All the references to sources classified as “Sangam” literature here refer to the editions published by the Saiva Siddhanta Works Publishing Society, Tirunelvēli. On Korkai, see also the epic Cilappatikāram 23.11; 27.127–35.

10 Pieris, P. S., “Nāgadīpa and Buddhist Remains in Jaffna,” JRAS, Ceylon Branch, XXVI (1917), 28.

11 Finot, Louis, Les Lapidaires Indiens (Paris, 1896), pp. xxxviii, xlv, xlviii, 64, 162.

12 Arthaśāstra ii.11 refers to gems called pārasamudraka. Such legends as Kalyāna-dhamma Jātaka (trans. Cowell, E. B., The Jātaka, II, 9091) may have grown from early gem trade with the island. Legends of the earliest stratum of Ceylon tradition several times refer to gems in connection with Naga kings and their thrones (Mahāvaṁsa 1.45, 48, 63, 75). On Aśoka's Rock Edict XII, see Law, B. C., “Historical and Geographical Aspects of the Aśokan Inscriptions,” Journal of Indian History, XLI, Part 2, p. 354.

13 Beal, Samual (trans.), Buddhist Records of the Western World (London, 1906), II, 239, 241, 243, 246.

14 Arthaśāstra ii.11. The promontory of Dhanuṣkoṭi was notable enough so that the Greeks knew of it as a destination and as a point from which to measure distances (Ptolemy i.13.1; vii.1.96). It may be suggested that Cūrṇī is not a river in Kēraḷa as pearls are not generally taken there, but may be Maṇṇār. Maṇṇāṛu means “muddy river,” it was a pearl market, and was located on the main ship channel. Pāśikā may refer to Pāśippaṭṭiṉam, an old town on Palk Bay, where pearls have been available on occasion.

15 Ibid, ii.11.115.

16 Strabo, , Geography xv.1.14–15.

17 Strabo xv.1.14; ii.1.4–7. Strabo himself refers to the people around Cape Comorin as Coniaci (XV.1.11).

18 Mahāvaṁsa, Chaps. 8–10.

19 Ibid. 7.38–42. All the earliest Greek sources refer to the island only as Taprobane, from the name of this early port and capital. Its association with the Vijaya myth is incorrect, for we show below that the Siṁhalas came later than the Paṇḍu princes.

20 Paranavitana, S., “Newly Discovered Historical documents Relating to Ceylon, India, and South-east Asia,” lecture reprint, given at Colombo, Dec. 4, 1964, p. 9; also, “An Account of Alexander the Great and Greek Culture in a Universal History Written in the Reign of Mahāsena,” lecture reprint, Oct. 31, 1964. Both these reprints, mimeographed by the Ceylon Department of Archaeology, contain a wealth of new data from epigraphical sources, which will, when completely edited and assimilated, demand much revision of the historiography of early Ceylon.

21 Some of the more important data are collected in Ray, H. C. (ed.), History of Ceylon (Colombo, 1959), Vol. I, Part 1, 89–111.

22 See Paranavitana's two lecture reprints. The credibility of these accounts is enhanced by the inclusion of a wealth of other data about the history and culture of the Greeks and their appearance in India.

23 Neither Greek nor North Indian sources before or during the Mauryan period refer to the island by any name except one derived from Tambapaṇṇi. Greek and Tamil sources from the first century on, however, use names derived from Siṁhala. The earliest recorded use of this name is a Tamil Brāhmī inscription of about the second century B.C. at Tirupparahkuṉṟam, which refers to flam (Sīhaḷa).

24 Arrian, 8.

25 Dīpavaṁsa 9.1–26; Mahāvaṁsa 6.5, 46.

26 Ibid. 8.6. Many other names in the earliest Ceylonese literary and epigraphical sources, such as Grāmaṇeyas, Kambojas, and Yonas, point back in the same direction.

27 Beal, , II, 67.

28 MBh Sabhā. 14, 30, 31, 32.

29 Epigraphia Zeylanica, V, Part 2, 219, 236, 237, 239, 251; note also such early names in the chronicles as Muṭasiva, Mahāsiva, Girikaṇḍasiva, Sivali, and (king) Siva.

30 Pliny, vi.22.

31 Strabo, xv.1.8.

32 Siculus, Diodorus, Bibliotheca Historica, ii.39; xvii.94.

33 Arrian, , Anabasis of Alexander, vi.20; Doidorus Siculus, xvii.104.

34 Paranavitana, S., “An Account of Alexander…,” p. 8.

35 Patirruppattu 11.1–5; Turanāṉūru 55.18–19; Akanāṉūru 266.20–21; Tirumurukārruppaḍai 125.

36 Paranavitana, S., “Pre-Buddhist Religious Beliefs in Ceylon,” JRAS, Ceylon Branch, XXXI (1929), 302–27.

37 Epigraphia Zeylanica, V, Part 2, pp. 210, 217, 218, 220, 231.

38 Mahāvaṁsa 16.12–14.

39 Paranavitana, S., “Inscriptions of Rājagala in Batticaloa District,” University of Ceylon Review, XX, No. 2 (1962), 159–62.

40 Few of the earliest Tamil Brāhmī inscriptions are published to date, though some may be seen in Epigraphy: Madras Reports, 1908, 1912, 1915, 1917, and 1918. The Department of Epigraphy has copies of others. Some of them around Madurai are easily accessible. See also Zvelebil, Kamil, “The Brahmi Hybrid Tamil Inscriptions,” Archiv Orientálni 32 (1964), 547–75.

41 Dani, A. H., Indian Paleography (Oxford, 1963), pp. 7274, 6668, 216–19, 225. This author, however, ignores the historical correspondences and works with paleography alone, and therefore his dating of all the Brāhmī scripts of the South is two to three centuries too late.

42 Epigraphia Zeylanica, V, Part 2, 227–34.

43 Nichols, C. W., “Text of the Brāhmī Inscriptions in the Ruhuṇa National Park, JRAS, Ceylon Branch, New Series, II, Part 2 (1952), 126–40.

44 Beal, , II, 239, 241, 243, 246.

45 Paranavitana, , “Newly Discovered …,” p. 9.

46 Ibid.; Mahāvaṁsa 13.11.

47 Most of these are datable to about the first century: Wheeler, Mortimer, “Arikamedu: an Indo-Roman Trading Station on the East Coast of India,” Ancient India, II (1946), 109. At least one, however, is said to have been recovered from the “megalithic” stratum: Casal, J. M., Fouilles de Virampatnam-Arikamedu (Paris, 1949), Annexe V, pp. 20, 55.

48 Some of the symbols in the inscription at Alakarmalai arc similar to those mixed with the earliest Brāhmī writing of Ceylon. See Epigraphia Zeylanica, I, p. 15; V, Part 2, pp. 229–33; Parker, H., Ancient Ceylon (London, 1909), pp. 438, 447. Similar symbols are also found on early punch-marked silver coins of both regions and on copper coins of the Jaffna Peninsula, and may be the type of writing called Kaṇṇeluttu in Cilappatikāram 26.136, 170.

49 Paranavitana, S., “Tamil Householders' Terrace, Anurādhapura,” JRAS, Ceylon Branch, XXXV (1940), 5456.

50 Epigraphia Zeylanica, V, Part 2, 242.

51 Mahāvaṁsa 21.10.

52 Ray, , p. 146.

53 Mahāvaṁsa 21.13–34; 33.56–61.

54 This inscription just south of Madurai can be seen written in a rock-cut cave above four sleeping platforms apparently carved out for four monks.

55 Iraiyaṉār Akapporuḷ (together with “Nakkīrar's” commentary) (Madras, 1939), pp. 5–7.

56 Mahāvaṁsa 7.50–58. Also, see Ray, , p. 94 on translation of the passage. This story is not found in the Dīpavaṁsa. Though “Vijaya” can hardly be considered a historical personage, the episodes relating to him demand careful scrutiny for the early history of the region.

57 Parañcōti Muṉivar, Tiruviḷaiyāḍal Purāṇam, I, Story 13.

58 The word Maturōtayanallūr, “eminent fertile town of Maturai,” appears in a fragment of a medieval inscription on a reused stone inside the door-way to the left, in a small temple standing between Koṟkai and the adjacent hamlet of Akkaśālai. The context of the word is land measurements, presumably local.

59 Arthaśāstra ii.11.2 speaks of Pāṇḍyakavāṭa and and Tāmraparṇī as sources of pearls. MBh, Droṇa 23, says that Kṛṣṇa overcame the Pāṇḍya kapāṭa and slew its king, and some versions refer to the place specifically as Kapāṭapuram. See Mu. Irākavaiyaṅgār, “Kapāṭapuramum Kaḍal Koḷum,” Annals of Oriental Research, II, Part 2, 19371938. Rāmāyaṇa vi.41, in narrating Hanuman's journey to the south, says he is instructed to cross the Kāvēri, then go south and cross the Tāmbaraparaṇi, and where that river “enters the sea is the golden gate of the Pāṇḍyas,” and beyond that is Mahendra (Potiyil hill, near Cape Comorin), and then there is Ceylon. This mention of the Pāṇḍiyan capital at the mouth of the Tāmbaraparaṇi must have been derived from a source older than this work as it now stands.

60 Puṟanāṉūru 9.8–11; Cilappatikāram, “Vēṉirkātai,” 1–2. The Paḥruḷi is associated with the Cape Comorin region, and also with South Madurai according to the legend of the Three Sangams. See Cilappatikāram 11.19–22, and also a legend about the inundation of the region of the Paḥruḷi in Rasanayagam, M., Ancient Jaffna (Madras, 1926), p. 10.

61 Paripāḍal 11.11–12 refers to an ascetic on Potiyil hill, which later became the sacred seat of Agastya. See also Puṟanāṉūru 33.7.

62 A summary of the historical import of some aspects of the Agastya legends is found in Sastri, K. A. Nilakanta, A History of South India (Oxford, 1958), pp. 6775.

63 Paranavitana, , “Newly Discovered…”p. 11.

64 Menon, K. P. Padmanabha, History of Kerala (Cochin, 1937), I, 22, 51; IV, 475 ff.

65 Nichols, C. W., “Brāhmaṇas in the Early Sinhalese Kingdom,” University of Ceylon Review, VIII, No. 4 (1950), 259–61; Mahāvaṁsa 7.44; 10.102.

66 Ptolemy vii.1.8, 74.

67 Puranāṇūru 343.1–10; Akanāṇūru 149.7–11; Periplus, 54, 56; these sources are also supported by the distribution and dating of Roman coins in Kēraēḷa.

68 Patirruppattu 17.4–6; 21.23; 22.15; 55.3–6; 67.1–4; 74.6; 76.4–5; 88.4–6; 90.19–30; patikams of 2nd and 5th decads; Kuruntokai 128.2; Puranāṉūru 343.1–10; Akanāṉūru 57.15; 127; 149.7–11, 199.19–24; Periplus 53, 54, 56; Ptolemy vii.1.7–8; Strabo ii.5.12; Pliny vi.23.26.

69 Puranāṉūru 66.1–3; 95; Perumpāṇārruppaḍai 28–38; Maṇimēkalai 25.124–26.

70 Hsüan-tsang said there was a stupa there built by Aśokarāja. Beal, II, 228–30.

71 Paṭṭiṉappālai 116–37; Ptolemy vii.1.13.

72 Cilappatikāram 14.104–12; Maṇimēkalai 25.124–26.


Altmetric attention score

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed