Pramāṇa · Search · About · API

Studies in the Geography of Ancient and Medieval India c10: VI. Bengal

Studies in the Geography of Ancient and Medieval India c10

Language
English

VI. Bengal

c10.1

European (mainly Portuguese) writers of the 16th and 17th centuries place a certain City of Bengala near the Meghnā estuary, *i.e.* the joint mouth of the Padmā and the Meghnā. This estuary extends over the wide area between the Districts of Buckergunge and Chittagong in East Pakistan. As Bengala (like the modern name *Bengal*) is a foreign corruption of *Vaṅgāla*, a celebrated historian has suggested¹ that this late medieval City of Bengala (which he locates near modern Chittagong) was the capital of the ancient Vaṅgāla-deśa and ‘gave its name to the kingdom, or vice versa, and, in either case, the old kingdom of Vaṅgāla must be located in the region round the city’.² He also suggests that the celebrated Pāla and Candra dynasties of Bengal originally ruled in the Chittagong region.³

c10.2

¹ See R. C. Majumdar in *I.H.Q.*, Vol. XVI, pp. 227-35.

c10.3

² *Ibid.*, p. 229 ; cf. p. 232, : “……this orginal kingdom must be in this region [of Chittagong and Dianga]”.

c10.4

³ *Ibid.*, p. 235.

c10.5

The above theories appear to be unwarranted. The City of Bengala, mentioned by foreign travellers in the late medieval period,⁴ seems to have nothing to do with the early medieval kingdom of Vaṅgāla, which originally denoted a smaller area, but whose geographical sense gradually expanded so as to include ultimately the whole of the land of the Bengali-speaking people.

c10.6

⁴ On the strength of Ibn Batūta’s reference to ‘Sudkawan (Sātgāon) and Bengala’ and to ‘Lakhnauti and Bengala’, it has been suggested that in these two early cases Bengala refers to the city and not to the country. There can be no doubt however that the names refer to provinces. According to Muslim authors both Rāl (Rādha) and Barind (Varendra) formed parts of Lakhnauti (Raychaudhuri, *Stud. Ind. Ant.*, p. 191). Similarly Bengala no doubt means the province of South-Eastern Bengal, sometimes also called Sonārgāon after its chief city. Orissa was usually called Jājnagar by Muslim authors. Such reference to Bihār, Lakhnauti, Bang and Jājnagar (*I. H. Q.*, *op. cit.*, p. 235) would, if we follow the said line of argument, suggest the existence also of a city called Bang. But this was never the case. Some later Muslim writers refer to East Bengal as *Bang* and to the whole of Bengal as *Bangāl*.

c10.7

In this connection, we have to trace the gradual expansion of the geographical connotation of the name Vaṅgāla. This name is often found in records and works of the period later than the 10th century A. D. Reference to Vaṅgāla is found in the Tirumalai inscription (1024 A. D.) of Rājendra Cola,¹ which speaks of king Govindacandra as the lord of Vaṅgāla-deśa. The Ablur inscription of Kalacurya Vijjala² (1157-67 A.D.) mentions Vaṅga and Vaṅgāla separately. This differentiation seems to be supported by a work called Ḍākārṇava which makes separate mention of Vaṅgāla and Harikela (=Vaṅga, according to the Abhidhānacintāmaṇi of the 12th century lexicographer Hemacandra),³ and by the Hammīramahākāvya of Nayacandra Sūri (15th century), which mentions Vaṅga and Vaṅgāla side by side.⁴ In the 16th century, however, Abul Fazl says, "The original name of Bangāl (=Vaṅgāla) was Bang (=Vaṅga). Its former rulers raised mounds measuring ten yards in height and twenty in breadth, throughout the province, which were called āl. From this suffix the name (i.e. Bangāl) took its rise and currency."⁵ This identification of Vaṅga and Vaṅgāla has to be reconciled with the evidence referring to them as two different countries.

c10.8

¹ Ep. Ind., Vol. IX, p. 229 ff. The earliest reference to Vaṅgāla has been recently traced in the Nesari plates (805 A.D.) of Rāṣṭrakūṭa Govinda III, which speak of Dharma (i.e., Dharmapāla) as the king of Vaṅgāla (J.A.S., Letters, Vol. XXII, 1956, pp. 133-34 ). This shows that the name Vaṅgāla was not unknown in earlier times and that the Pālas originally ruled in the tract called by that name. It is interesting to note that Dharmapāla is apparently called 'lord of Vaṅga' in the Sāgar Tāl inscription of Bhoja and 'king of Gauḍa' in the Sanjan plates of Amoghavarṣa. With the expansion of their territories, the Pālas preferred to style themselves as 'lords of Gauḍa'. This seems to be the reason why Vaṅgāla did not become popular in the sense of the Pāla empire.

c10.9

² Ep. Ind., Vol. V, p. 57.

c10.10

³ Raychaudhuri, op. cit., p. 189.

c10.11

⁴ I.H.Q., op. cit., p. 237.

c10.12

⁵ Jarrett, Ain-i-Akbari, trans., II, p. 120.

c10.13

The Raghuvamśa reference to the defeat of the Vaṅga people in the land watered by the lower streams of the Ganges⁶ and epigraphic references to Vaṅga comprising the Vikrampur region of Dacca and Faridpur¹ and to the *Nāvya* region of Vaṅga very probably in the Faridpur and Buckergunge Districts,² leave hardly any doubt that Vaṅga certainly included at least parts of the present Dacca, Faridpur and Buckergunge Districts. In the lower part of this region very high roads or earthen-embankments are constructed even today in order to prevent the tide of floods and to facilitate communications during the rainy season. It may therefore be suggested that the southern part of old Vaṅga thus at first came to be known as Vaṅgāla. It is interesting to note that this view is supported by other evidences.

c10.14

⁶ वङ्गान्युत्खाय तरसा नेता नौसाधनोद्यतान् । निक्खान जयस्तम्भान्गङ्गास्रोतोन्तरेषु सः ।। (Raghuvamśa IV, 36) Mallinātha explains Gaṅgā-sroto-natreṣu as Gaṅgāyāḥ srotasāṃ pravāhānām=antareṣu dvīpeṣu; cf. gīt-āntareṣu, ‘in the intervals of singing’, vāṣpa-salil-āntareṣu, ‘in the interval of weeping’; na mṛṇāla-sūtram racitaṃ stan-āntare, ‘……in the space between the breasts’, etc. See below, pp. 139-40; Ch. XI (p. 172 ff.)

c10.15

¹ Cf. *Vaṅge Vikramapura-bhāge*, etc. (*Ins. Beng.*, Vol. III, p. 125).

c10.16

² Cf. *Vaṅge Nāvya Rāmasiddhi-pāṭake*, etc., *Nāvya Vinayatilaka-grāme pūrve samudraḥ sīmā*, etc. (ibid., p.146). *Nāvya* means a region accessible by boats. That this region was not far from the sea (*samudra*), *i.e.* the estuary, is also clear. The word means apparently the same thing as Bengali *bhāṭi*, ‘with the tide, the region towards the sea’ (from *bhāṭā*, floods of the tide running towards the sea). Cf. also East Bengali *bhar*, low region. In the Muslim period, the district called *Bhāṭi* sometimes indicated the coastal strip between the estuaries of the Hooghly and the Meghnā. The late work called *Śaktisaṅgama Tantra* places Vaṅga-deśa between the *ratnākara* (sea) and the Brahmaputra (above, p. 84).

c10.17

As said above, the Tirumalai inscription refers to Govindacandra as the lord of Vaṅgāla-deśa. The inscriptions of the Candra dynasty of East Bengal, to which Govindacandra belonged, however say that Trailokyacandra, the first king of the family, became lord of Candra-dvīpa and flourished as the mainstay of the fortune (*i.e.* as a feudatory) of the king of Harikela (=Vaṅga according to Hemacandra’s *Abhidhānacintāmaṇi*).³ Thus the Candra kingdom is called Candra-dvīpa according to one evidence and Vaṅgāla-deśa according to another. Candra-dvīpa and Vaṅgāla-deśa thus appear to be more or less identical. As Candra-dvīpa is no other than the celebrated Bāklā-Candra-dvīp (*i.e.* parts of the Buckergunge District and the adjoining region), the Buckergunge area was apparently included in Vaṅgāla-deśa.

c10.18

³ *Ind. Cult.*, VII, p. 411. According to Keśava’s *Kalpadrukośa*, Harikela or Harikeli was the name of the Sylhet (*Śrīhaṭṭa*) region. Cf.

c10.19

श्रीहट्ट हरिकेलिः स्याच्छ्रीहटोप क्वचिद् भवेत् ।

c10.20

This name came to be applied to Vaṅga apparently due to the temporary annexation of Vaṅga to the Harikela kingdom. Cf. *Ep. Ind.*, Vol.XXVIII, pp. 338-39.

c10.21

The suggestion is further supported by the identification of Vaṅgālavaḍā-bhū in the Rāmasiddhi-pāṭaka of the *Navya* region of Vaṅga, mentioned in the Sāhitya Pariṣad plate¹ of Viśvarūpa-sena, with modern Bāṅgroṛa in the region of Rāmsiddhi in the Gaurnaḍi region of the Buckergunge District. Vaṅgāla thus appears to have been popularly mentioned as a separate political unit, after the Candras had established a kingdom in Vaṅgāla; i.e. the southern part of Vaṅga. Śrīcandra, son of Trailokya-candra, conquered the Dacca and Faridpur Districts, as the find-spots of his records would show. Laḍahacandra, grandson of Śrīcandra, apparently ruled also over the Tipperah District. Thus Vaṅga proper was now included in the new kingdom of Vaṅgāla. As a result of this, the name Vaṅgāla could be optionally used in an expanded sense to indicate a large area in East Bengal that formed part of the kingdom of Śrīcandra and his successors. By this time, Vaṅga and Vaṅgāla apparently signified more or less the same territory.

c10.22

¹ N. G. Majumdar, *Ins. Beng.*, Vol. III, pp. 840 ff.

c10.23

Early Muslim authors refer to the Muhammadan kingdom of Bengal first as Lakhnautī or Gaur, as their first settlement did not include East Bengal. The kingdom of East Bengal was sometimes referred to as Bang, Bangāl or Sonārgāon; and sometimes the whole of Bengal was mentioned as Gaur-Bangāl.² Sātgaon was often recognised as a separate political unit. But, as early as the reign of Ghiyāsuddin Tughluq Shāh (1320-25 A.D.), some Muslim authors are known to speak of Bangāl in a wide sense so as to include Lakhnautī, Sonārgāon and Sātgaon.³ In the 16th century, the name Bangāl loosely indicated a wider area. It may or may not have geographically included the Chittagong region which was included in the Mughal Subah of Bangāl only during the reign of Aurangzib in the 17th century. There is no proof at all that the Chittagong area formed part of the old kingdom of the Candras and the Pālas. It must be noted that not a single inscription of those dynasties has so far been discovered in the Chittagong District.¹

c10.24

² The *Śaktisaṅgama Tantra* divides Bengal into two halves, viz. Vaṅga and Gauḍa (lying between Vaṅga-deśa and Bhuvaneśa or Bhubaneswar). See above, pp. 84, 98.

c10.25

³ Raychaudhuri, *op. cit.*, p. 191; Raverty, *Ṭabaqāt-i-Nāṣirī*, p. 590n. Mulsim authors popularised the use of *Bangāl* in preference to other names to denote the whole of Bengal. A Nepal inscription of 1346 A. D. refers to the army of Sulṭān Shamsuddin Ilyas as *Vaṅgāla-bahula bala* (*J.B.O.R.S.*, Vol. XXII, p. 81; *I.H.Q.*, *op. cit.*, p. 227).

c10.26

¹ The Arab merchant Sulaiman (851 A.D.) refers to the powerful kingdom of Rahma which has been identified with the kingdom of the Pālas and the identification of Rahma, Rahmi and Ruhmi of the Arabs with Rāmu, a few miles east of Cox's Bazar, has been upheld (*I.H.Q.*, *op. cit.*, pp. 232-34). Apart from the fact that there is absolutely no proof of Pāla occupation of Ramu, it seems unconvincing that the Bihar-Bengal empire of the Pālas (even if it included Rāmu), with the emperors often residing at *jayaskandhāvāras* in Bihar, would be named after Rāmu which probably formed a part of Arakan. That the said theory is unjustifiable is proved by Ibn al Fakih (902 A.D.) who says, "In India lies a realm called Rahma bordering on the sea. *Its ruler is a woman*. It is ravaged by the plague and any man who comes from elsewhere in India and enters the country (Rahma) dies there. Yet many come by reason of great profits to be made." (Harvey, *Hist. Burma*, p. 10). In 902 A.D., the Pāla empire is definitely known to have been under Nārāyaṇapāla (c. 857-911 A.D.) and not under a woman. Rahma may be the kingdom of Thaton called Ramañña-desa (Lower Burma). Marco Polo who placed Bangala 'tolerably close to India' and under the rule of the king of Mien (Burma), and Fakir Muhammad who placed Bangāl to the east of Bhāṭi apparently made some confusion. The Bengali ballad called *Māṇikcandra-rājār Gān*, which says *Bhāṭi haite āila Vāngāl lambā lambā dāṛi*, 'the Vāngāls (people of Vāngāla proper) with their long beards came from Bhāṭi (the southern country)', proves that the people of East Bengal, who were responsible for the original composition of the ballad, knew Vāngāla to be identical with Bhāṭi (see C. C. Banerji, *Caṇḍīmaṅgala-bodhinī*, II, p. 765). Mukundarāma (16th century) in his *Caṇḍīmaṅgala* (C.U. ed., p. 655) speaks of the Vāngāls as good sailors. As the Chittagong people are reputed sailors, it may be suggested that Mukundarāma identifies the land of the Vāngāls with Chittagong. But Mukundarāma was a man of West Bengal and to the people of his part of the country, the inhabitants of any District of East Bengal are Vāngāls even today. Moreover, the passages in question are certainly interpolated in the *Caṇḍīmaṅgala*. The *Māṇikcandra-rājār Gān*, an East Bengal production, is therefore more important in this connection. There in we see that the people of Bhāṭi were known even to the people of East Bengal as Vāngāls, *i.e.* inhabitants of Vāngāla.

c10.27

We now come to the City of Bengala mentioned by the European (mainly Portuguese) writers of the 16th and 17th centuries A.D. Duarte de Barbosa, Barthema (1510 A.D.), Ovington (1639 A.D.) and several other writers refer to the city. Some authorities identify the city with modern Chittagong or locate it near about that place. It is shown in a map drawn by Gastaldi in 1651 and in another in the *Travels* of Cornelius le Bruyan (published in 1701). In these maps and works as well as in those of Blaev, Sausson, Purchas and others, the City of Bengala is indicated in the Chittagong region.² But Barbosa, one of the earliest Portuguese writers on Indian geography, says that the Bay of Bengal is ‘a gulf which enters towards the north and at its inner extremity there is a great city inhabited by Moors (Muhammadans), which is called Bengala, with a very good harbour.’¹ The expression ‘at the inner extremity’ may point to a locality far to the north of Chittagong, and Moreland suggests that the Portuguese meant Sonārgāon by the City of Bengala.² But the city apparently belonged to what the Portuguese called the Porto Grande = Portus Mangnus = Great Port, and we have to determine the location of the latter.

c10.28

² Raychaudhuri, *op. cit.*, p. 189; *I.H.Q.*, *op. cit.*, p. 229n.

c10.29

¹ I.H.Q., op. cit., pp. 229-30.

c10.30

² India at the Death of Akbar, p. 309.

c10.31

Portuguese writers, when they speak of a voyage to Bengal, usually refer to the Great Port as well as the Porto Pequeno = Portus Parvus = Little Port. Portuguese *porto* is usually understood in the sense of English *port*, and the Little Port is identified with Hooghly or Sātgāon while the Great port is located near Chittagong. Moreland has however shown that ‘the word *porto* in the mouth of the sea-faring Portuguese at the period referred primarily to a gap in the coast line and not, as landsmen are apt to assume, a town on the sea-shore; in other words, *porto* might signify a gulf or estuary which might contain several sea ports.’³ He has pointed out that sometimes the Little Port indicated the Hooghly estuary, and the Great Port meant the Meghnā estuary and included both Chittagong and Śrīpur (in the Dacca District). The location of the Great Port therefore does not help us materially in locating the City of Bengala.

c10.32

³ Ibid., 307-08. Father F. Fernandus, the Jesuit missionary who was sent to Bengal in 1598, speaks of the danger when his ship ran aground in the Portus Parvus; but it was afloat and, after sailing for eight days within the Portus, it reached the Portuguese station of Hooghly. By *Potrus Parvus* therefore Fernandus meant the Hooghly river and not any port. He next went to the Portus Magnus and first reached Śrīpur which he describes as a station in the Portus Magnus. Śrīpur was an important port on the Padmā about 18 miles from Sonāragāon, the eastern capital of Bengal. The Father then arrived at Chittagong which was also a station belonging to the Portus Magnus. According to Fernandus therefore the Great Port extended from the Karṇaphuli river to the immediate neighbourhood of Dacca. By the word *porto* some writers however actually meant a port; Fitch, *e. g.*, identifies Porto Pequeno with Sātgāon. According to Moreland, the change in the meaning may be attributed to the fact that ‘the Portuguese did not to any great extent trade directly with Śrīpur; their communications were with either Hooghly or Chittagong, that is, with only one *Station* in each *Porto*, and, in these circumstances, the transfer of the name from the *Porto* to the *Station* might easily take place, just as the *Mersey* has become a synonym for Liverpool’ (*op. cit.*, p. 309).

c10.33

The chief points of interest in this problem are two. Firstly, there is not a single reference to this City of Bengala, so famous to the Portuguese, in the whole range of medieval Bengali literature, not even in the Bengali history of the Tripurā royal family which often mentions Chittagong. The Bengalis therefore did not know any place of that name, or knew it by a different name, i.e. a name of their own. Secondly, while some European writers like Rennell could not trace the city or its site, others denied the very existence of the city called Bengala. In 1689, Ovington. remarked, “A late French geographer (Baudrand) has put Bengala into his Catalogue of imaginary cities, and such as have no real existence in the world.” It therefore seems quite clear that even Europeans other than the Portuguese sometimes considered the name of the City of Bengala as a misnomer and had no memory of it after the collapse of the Portuguese power in Bengal. The City of Bengala thus appears to be a name given by the Portuguese to a city which was usually known to others by its Bengali name. As however there is absolutely no proof (in the Bengali literature and Bengal records of any period) of the existence of any city called Vaṅgāla in the south-eastern part of Bengal, the Portuguese name *City of Bengala* seems to have originally indicated the city *par excellence* (i.e. the chief city) of the country of Vaṅgāla, i.e. East Bengal. Names like Andhra-pura, Magadhapura or Māgadhapura, etc., are known to have been alternative names of the chief cities derived from the names of countries, and we have seen that the south-eastern, north-western and south-western provinces of Bengal were also known to the Muslim writers as Sonārgāon (Vaṅgāla), Lakhnautī (Gauḍa) and Sātgāon (Rāḍha) respectively after the chief cities of those provinces. As the chief city of East Bengal lay not far from the Meghnā estuary,² it might have been called the City of Bengala by the Portuguese. But apparently that name was used by foreigners (chiefly Portuguese), and the Bengalis continued to use the local or Bengali name. That is why the name could not be traced after the Portuguese collapse in Bengal.¹

c10.34

¹ Bengal Past and Present, Vol. XIII, p. 262 ; I.H.Q., op. cit., p. 230n.

c10.35

² Sonārgāon was the capital of East Bengal during early Muslim days. Dacca was made the capital of the Bengal Subah early in the 17th century. The remark of Purchas that ‘Gauro (Gaur or Lakhnauti), the seat royal, and Bengala are fair cities’ (I.H.Q., op. cit, p.230) apparently refers to the fact that Bengala was the chief city of East Bengal and was looked upon as a secondary capital of Bengal.

c10.36

¹ Quite different is the case with Meghnā (pronounced Māgnā in East Bengal, with the first vowel sounding like, *a* in *man*) apparently derived from the Latin word meaning 'great', which has supplanted the older name of the river.

c10.37

It must however be admitted that some authorities locate the City of Bengala near about Chittagong.² It seems that originally the chief city of East Bengal was known to the Portuguese as the City of Bengala; but after the Portuguese port in the Chittagong region had become the most flourishing centre of trade in Bengal, they began to call it the City of Bengala. This again seems to have been done when Sonārgāon passed out of the picture due to the transference of the provincial capital first to Rājmahal and then to Dacca. The fact that Portuguese sailors came by sea and first reached this flourishing port after days of hardship in the sea may also have emphasized its importance to them as the city *par excellence* of the country of Bengal.

c10.38

² It has been pointed out that Ortelius places Bengala in the same place where Hommanus places Chatigam or Chittagong. Yule refers to a chart of 1743 and concludes that Chittagong is probably identical with the City of Bengala. Campos also identifies Chittagong and Bengala. Ovington places Bengala to the south of Chittagong. In Blaev's map and Sausson's chart the city is located on the southern bank of the Karṇaphuli near about the place where Broucke places Dianga. This location is supported by Vignorla's map of 1683. But in an old map in Thevenot, the city is placed above Xatigan or Chittagong. Hosten and Majumdar think that Dianga, opposite Chittagong, represents the site of Bengala. It is also suggested that originally both Dianga and Chittagong were included in the city. Hosten identifies Dianga with a place now called Bandar on the left bank and near the mouth of the Karṇaphuli river. He also takes Dianga to have been the Porto Grande. Blochman identifies Dianga with Dakhiṇ-ḍāṅgā or Brāhman-ḍāṅgā bath on the Sāṅgu river to the south of Chittagong (I.H.Q., *op. cit.*, pp. 231-32 and notes). Blochman may be right, as the foreign name seems to correspond to the Bengali word *ḍāṅgā* which is pronounced in East Bengal as *ḍāṅgā*, with the first vowel sounding like the English *a* in *man*.

c10.39

Whatever however the location of the City of Bengala may be, there is no doubt that it had nothing to do with the old country of Vaṅgāla known to have been the kingdom of the Candra kings of East Bengal.³

c10.40

³ See aslo S. H. Hodivala in *Num. Supl.*, No. XXXIV (*J.A.S.B.*), pp. 199-121. *Cf. Ind. Cult.*, Vol. VII, pp. 407-08.

c10.41

There has been some discussion on the derivation of the name *Bengal*.¹ But the conclusions are not satisfactory. Since the emergence of modern Hindi from its Apabhraṃśa stage, the speakers of Hindi and other languages have been calling the Bengali-speaking area by the name *Vaṅgāl* (without the final *a* in the last consonant), which has been transliterated by the English in their script as *Bengal*. The Muslims first came to India when the final *a* in the last consonant of Sanskritic words in North Indian dialects was pronounced. The earlier pronunciation of *Vaṅgāl*, as is well known, was *Vaṅgāla* which was naturally transliterated by the Muslims in their script as *Bangālah* (pronouncing *Bangāla*). This Muslim pronunciation of the name is directly responsible for the name *Bāngālā* now applied by the Bengalis to their country. The first *ā* in the name is comparable to the first *ā* in Bengali words like *hājār* (thousand) derived from Persian *hazār*. The change of *a* into *ā* is due to the fact that *a* pronounced outside East India resembles *ā* as pronounced by the Bengalis. The last *ā* in the name is again comparable to Bengali words like *khānā* derived from *khānah*.² The people of Bengal is known elsewhere in India as *Vaṅgāli* which the English transliterated as *Bengali* or *Bengalee* and which the Bengalis have made *Bāngāli* in their language. This word is also influenced by such Muslim words as *Kābuli*.³ It is interesting to note in this connection that, according to a general ethnological principle, the specific name of a tribe often originates among neighbouring tribes and is eventually adopted by the tribe to which it is applied.

c10.42

¹ See N. N. Das Gupta in *I.H.Q.*, Vol. XXII, pp. 277-85.

c10.43

² Cf. also *bandah* (slave) = Bengali *bāndā*, *khajānah* (revenue) = Bengali *khājānā*, etc. Just as the word orginally spelt and pronounced in Bengali as *khājānā* has now been modified to *khājnā* (with the elision of the central *ā*), the earlier Bengali spelling and pronunciation of the name *Bāngālā* have now given way to the modified form *Bānglā* (sometimes softened into *Bāṅlā*). That, during Muslim rule, the Bengalis adopted the Muslim pronunciation of the name of their country is probably indicated by the fact that today Bengalis of any education speak, in their common speech, usually of the English *India*, *Bengal* and *Calcutta* in preference respectively to the names *Bhāratavarṣa*, *Bāngālā* (or *Bānglā* or *Bāṅlā*) and *Kalikātā*. Cf. also the popularity of the Anglicised surnames Banerji, Mukherji, etc. The introduction of a large number of Persian and Arabic words in the vocabulary during Muslim rule has further to be compared with the later adoption of numerous English words during the British period.

c10.44

³ Cf. the *yā-yi nisbati* in Persian and the corresponding Arabic suffix to from relatives (D. C. Phillott, *Higher Persian Grammar*, pp. 400 and 714-17).

c10.45

To sum up the discussion, the Candra kings of South-East Bengal are sometimes represented as lords of Candra-dvīpa (Bāklā-Candradvīp in the Buckergunge region) and sometimes as lords of Vaṅgāla-deśa. This fact points to the location of Vaṅgāla in its original geographical sense in the coastal region of Southern Bengal. The connotation of the name began to expand with the expansion of the Candra kingdom of Vaṅgāla over wide areas of South-East Bengal. The name was popularised in the sense first of East Bengal, and then in that of the whole of Bengal, by the Muslims. The location of Vaṅgāla proper in the Buckergunge region near the Bay of Bengal, that is to say, in the southern part of the ancient Vaṅga country, seems to be supported by Abul Fazl’s identification of Vaṅga and Vaṅgāla and his derivation of the latter name from Vaṅga+āl (Sanskrit āli) on the grounds that ‘its former rulers raised mounds measuring ten yards in height and twenty in breadth throughout the province, which were called āl.’ Although the derivation may be wrong and Vaṅgāla may have sprung from Vaṅga+the Prakrit suffix āla in the sense of a notable district belonging to Vaṅga,¹ we have seen that the interesting mention of the earthen mounds, primarily meant for keeping off the encroachment of sea-water from the corn-fields, refers to a condition prevailing in the Buckergunge region of the coastal area of Bengal even to-day. The name Vaṅgāla is very rarely found in records earlier than the rise of the Candras in the tenth century A.D. The mention of the name in the Goharwa inscription of Karṇa (1041-71 A.D.) in the description of the exploits of his predecessor Lakṣmaṇarāja proves its importance in the eleventh century. Of course Lakṣmaṇarāja, who is represented in the Goharwa inscription as having defeated the Vaṅgāla country or people, ruled about the middle of the tenth century when the Candra power had already been established in Vaṅgāla.²

c10.46

¹ Cf. my *Grammar of the Prakrit Language*, p. 33.

c10.47

² As has already been indicated above, the Nesari plates of 805 A.D. suggest that the name Vaṅgāla (apparently indicating the original territory

c10.48

As we have seen, the most important point in regard to the supposed existence of a ‘city of Bengala’ is that no such name of a great city is found in any of the numerous works of medieval Bengali literature. This question, usually ignored, has to be answered. Our answer to this question fully tallies with that of Hodivala who reached the same conclusion after throughly ransacking the available Muslim sources. In an attempt to locate the Mughal mint-town of Bangāla in Akbar’s empire, he says, ‘Briefly, there would appear to be *fairly good grounds* for thinking that *Bangāla* was not the real or fixed name of any town or city, but an alternative or honorific designation by which the *capital of the province at the time being* was known. Thus the *Bangāla* of Mun’im Khān’s time *might have been* Gaur, and it is not impossible that during the subsequent twenty years the name was sometimes applied to Tānda. The *Bangāla* of the coins of the 39th and following years of Akbar’s reign would, by parity of reasoning, be Akbarnagar.’”¹

c10.49

¹ See Hodivala, *op. cit.*, p. 211. It will be seen that the mint-town of Bangāla, whence Akbar’s coins were issued, could not have been situated in the Chittagong region (where the ‘city of Bengala’ is usually located), as that area lay outside the Mughal empire during Akbar’s rule. See also A. Cortesao in *J R.A.S.B.*, Letters, Vol. XI, pp. 10-14; cf. A.B.M. Habibullah, *ibid.*, pp. 33 ff.

c10.50

of the Pālas in the Buckergunge region) was known even in earlier times, though it does not appear to have been popular in the sense of the eastern part or the whole of Bengal before the rise of the Candras.

Linked texts (11)