The Concept of Bharatavarsa PDF

# The Concept of Bhāratavarsa and Its Historiographical Implications The choice of the concept of Bhāratavarsa as the theme of this essay has derived from a number of reasons. We have been brought up from our early childhood on the idea that the country we live in is Bhāratavarșa which is India and...

# The Concept of Bhāratavarsa and Its Historiographical Implications The choice of the concept of Bhāratavarsa as the theme of this essay has derived from a number of reasons. We have been brought up from our early childhood on the idea that the country we live in is Bhāratavarșa which is India and which is also a map with specific boundaries, separated from other countries with similar maps indicating them. The partition of the India of 1947 changed the map, but the notion of Bhāratavarsa and the name remained, conveying, as it did to our predecessors, the image of a country which has forever been there and will so remain despite the change in the map. And yet, the question of the history of India or Bhāratavarsa as it evolved over time, and linked to what is perceived as India today, remains to be critically examined in terms of historical change. In other words, the link between a notion or a concept of space, the actual geographical space supposed to be denoted by it, and the space as the locus of our history is an issue which needs to be reopened, because what we accept today as granted is based on a number of assumptions. These assumptions, without adequate deference to the many meanings embedded in our sources, have substantially affected our generalizations about Indian history, particularly of its early phase. One major assumption, for example, has been that of the identity of the concept of India with the concept of Bhāratavarșa. It is not possible, in this essay, to historically explain in what ways the convergence of the meanings of the notions took place, but it seems obvious that by the nineteenth century, whether in history-writing or in general thinking, their identity had been established. Those who write on India, or on the idea of it, take it for granted that what they mean is represented by the term Bhāratavarșa as well, and that they both carry with them the sense of our past or our history. Even in the early phase of colonial history-writing, it was easy to conceive of a History of India, and a corresponding indigenous enterprise in that direction would have produced, for example in a language like Bengali, a title like Bharatavarser Itihās. An An academic example of the identification of Bhāratavarsa with India is a positive statement by a reputed researcher of Purāņic cosmography, who wrote: > The southernmost varsa, Bhārata, lying between the Himavat and the sea, is, of course, India. *(emphasis added)* The understanding and unhesitating acceptance of the identity of India and Bhāratavarşa was further formalised in the solemn declaration of our constitution: 'India that is Bharat shall be a union of States.' This declaration puts a historic seal on the identity of our country and nationality, but not necessarily on our history. The terms, it needs to be remembered, had different origins, one perceiving the country from what may be called a geographically outer perspective and conveying different meanings in different contexts, and the other term, Bhāratavarsa, consistently, but not eternally, used in early texts of different varieties, located within a completely different cosmographic structure. The term Bhāratavarşa has therefore altogether different nuances, and the texts present variations on how its different segments are conceived. Pursuing the early history of this term, independently of its possible correspondences, and clarifying the different contexts in which they occur, may yield rewarding results. Connected with this issue is also the nature of historiography. We have been used for long to take it for granted, despite some recent efforts to explore the history of the idea of India, that the country we inhabit has had the same connotation all along, and that the way we think of our country now is what was always perceived in the past. At the same time, it is *common knowledge that geographical spaces and notions such as that of Bhāratavarșa are defined and redefined, and that, in order to understand the history of a space and its peoples, it is necessary to be aware of such processes of definition and redefinition. Secondly, identification of a particular collective sensitivity, which is usually termed nationalism, with a space is not a given quality of that space or of the collective human entity inhabiting that space. It is a sensitivity which is historically acquired and which may undergo mutations. A country or ideas about that country may exist independently of that collective sensitivity unless this historically acquired awareness is shown to be evident through different forms of articulation.* Today, when we have come to accept that a geographically bounded (in whatever way) and a constitutionally defined country is what we belong to, the need to look into the meanings of that country in the past seems to me, for various reasons, to be urgent. Historiographically, we are at a particular juncture in our efforts to understand that meaning, particularly because there are sharply different approaches to the concept of India or 'Bhāratavarsa'. Without getting into any great details one can perhaps locate three major positions in recent writings on the theme. One position, which seems to have taken off from the colonial construct of India as a territorial, governable unit, separable for administrative purposes from other spaces, insists on the idea of India or Bhāratavarșa as an expression of national unity present in the distant past. Monographs such as The Fundamental Unity of India, published in the early second decade of the previous century, forcefully projected this idea of unity; in this idea, 'unity' is a fundamental quality of the country, evidence of which could be located in concepts of Indian geography, networks of pilgrimage centres, expressions of urge for political unity through conquests or colonisation and so on. on. The The notion of the existence of our unified country has permeated writings on the history of India as one unquestionable given unit, despite the vitality of its regions, and the projection of the Indian nation in the past in relation to the geography of Bhāratavarsa is a refrain which continues to this day. Bhāratavarsa or India, as locus of an ancient nation, implied in The Fundamental Unity of India, is present in a recent work on The Concept of India as well, which suggests: > Obviously the inhabitants of the subcontinent (the country lying between the Himalayas in the north and the ocean in the south] were considered by the Puranic authors as forming a nation, at least geographically and culturally. There were feelings among at least a section of the public that the whole of the subcontinent (or by and large a major part of it) was inhabited by a people or group of peoples sharing a link culture or some common features of an 'umbrella' culture in so deep a manner that they could be called by a common name-Bhārati. So geographically and culturally, if not politically and ethnically, the Bhārati were a nation. An exactly opposite position seems to be taken in an essay, 'The Imaginary Institution of India', published in one of the volumes of the Subaltern Studies series. The essay opens, with a good deal of emphasis, with the following statement: > India, the objective reality of today's history, whose objectivity is tangible for people to preserve, to destroy, to uphold, to construct and dismember, the reality taken for granted in all attempts in favour and against, is not an object of discovery but of invention. It was historically instituted by the nationalist imagination of the nineteenth century. *(Emphasis added)* Apart from the consideration that the exercise behind this statement is not grounded on the use of any substantive historical documentation, the approach in the essay itself involves certain implicit assumptions which are open to questioning: (i) the essential equation that it posits between nationalism as 'historical reality' and the idea of India, (ii) 'invention' out of nothingness without any pre-existing concepts or notions which may have been 'objective reality' of a different kind, not necessarily denoting 'nationality', and (iii) attribution of the 'invention' to the nationalist imagination, ignoring the possibility of the emergence of the modern notion of India as a colonial space, and the relationship of that emergence with the construction of a particular state and its history. In denying the pre-national existence altogether of the institution without actually defining what an institution is, the essay seems to be denying the idea of India too, because it equates the 'objective reality of India' with the reality of Indian nationalism which is modern. This denial seems to be present in C.A. Bayly's Empire and Information too, in which the concept of India is seen as an important 'aspect of emerging national consciousness', geography, as a social science, being at the same time, 'close to the heart of British colonial information collection'. Bayly underlines the distinction between European and what he calls 'Hindu conceptualization of geography', and in the light of this contrast, characterises Bhāratavarsa as corresponding to Hindu 'sacred' space. The mapping of India, in Mathew Edney's view, was for the first time 'a massive intellectual campaign to transform a land of incomprehensible spectacles *(emphasis added)* into an empire of knowledge... the geographers created and defined the spatial image of the company's empire', and also its territorial integrity and its basic existence. Edney of course makes it clear that it was not a value-free space; the way the British 'represented India' made it their 'India'. This was 'British India' which comprised only what they perceived and governed. What, then, about pre-colonial times? There is apparently a position somewhere in between. Irfan Habib, for example, has been arguing strongly for the existence of the concept of India not only as a geographical unit but also as representation of a country in which certain special social and religious institutions are present as early as the fourth century BC, separating the country from others. The geographical and cultural separatedness of India was the basis, in Habib's argument, of Al Biruni's comprehension of India as a 'cultural unity', and, reinforced by long process of interaction and adjustment, 'some prerequirements of nationhood had... seemingly been achieved by the time the British conquests began in 1757... India was not only a geographical expression, it was also seen as a cultural entity and a political unit'. Without pausing to reflect on why such an overripe field had to wait, to follow Habib's own position, the colonial intervention for mature nationalism, what appears to be significant is this. In most although not all discussions on the idea of India, or on Bhārata-varsa, the issue of the nation and of nationalism somehow creeps in. This intermeshing is perhaps understandable but not inevitable, and in choosing to write on the theme of Bhāratavarsa, my idea has been to understand it as a historically evolved concept, to probe into the kind of senses in which the notion of Bharatavarșa was articulated by those who referred to it in different contexts, and on the basis of this probe delve into the possible implications of the concept for the historiography of early India. ## I. From Jana to Janapada In pursuing the concept of Bharatavarșa in diverse sources one must remember that there was a textual phase in early India in which the term Bhāratavarsa, even in a geographical sense, did not appear at all. In fact, early textual references were to janas, or peoples or communities, and to natural landmarks such as rivers by which locations of janas were defined. Thus, although the Bharatas are mentioned along with other janas in the Rgveda, they do not, like the others, figure in the contexts of fixed territories. It is in the Brāhmaṇa category of Vedic texts that various spatial directions, in relation to what was considered a central zone, are for the first time mentioned. Janapada as an inhabited country or the space where a jana resided figures also for the first time in such Brāhmaṇa texts as Taittiriya Brāhmaṇa, Aitareya Brāhmaṇa, Śatapatha Brāhmaṇa, and so on. The significance of dis or direction, which I shall point out later, is in relation to the janapadas, as dis, defined how the janapadas were to be located. Thus, this is how the Aitareya Brāh-maņa specifies different regions and those who inhabited and ruled over those regions; this invocation occurs in the context of- the Mahābhişeka of Indra: '... in this eastern quarter (Prācyānām disi), whatever kings there are of the eastern peoples (prācyānām rājānah), they are anointed for overlordship;... Therefore, in this southern quarter, whatever kings there are of the Satvants (daksi-ņasyām disi), they are anointed for paramount rule; ... in the western quarter (pratīcyām disi), whatever kings there are of the southern and western peoples, they are anointed for self-rule ... in this northern quarter (Udīcyām disi), the lands (janapadāh) of the Uttara-Kurus and the Uttara Madras, beyond the Himavat, their kings are anointed for sovereignty... in this firm middle established quarters (dhrūvāyām madhyamāyām pratisthāyām disi), whatever kings those are of the Kuru-Pāñcālas with the Vasas and the Uśinaras, they are anointed for kingship...' Clearly, in the enumeration of the directions of earthly kings, in the context of the great consecration of god Indra, the composer of the Brahmana shows greater familiarity with the janapadas of the firmly established middle region (madhyamā dis) than with those of other quarters. In the later Brahmanical discourse on the configuration of various janapada regions, it was this middle region which came to be regarded as the core or the centre from which other quarters or directions (dis) were taken to have radiated. The idea of a country, or rather a segment of earthly space, which accommodated the janapadas placed in different directions was for the first time articulated in the reference to Jambudvīpa which occurs in early Buddhist texts. One of four mahādipas (mahādvīpas), or four great islands, it extended around Mt. Sineru and was ruled by a cakkavatti or a sovereign ruler. In fact, according to the texts, it was only in Jambudvīpa that Buddhas and cakkavattis were born. When Metteya Buddha (Maitreya Buddha) appeared on earth, it was full of people and there were eighty-four thousand cities in it. According to Malalasekera, the author of the Dictionary of Pali Proper Names, when seen as different from Sihaladīpa or Tambapanņidīpa, 'Jambudīpa indicates the continent of India'. However, if one goes by a reference in the Anguttara-nikāya, there is a Jambudvīpa in each cakravāla or horizon, making it difficult to take Jambudvipa to correspond to the geography of any specific country like India. The concept of Jambudvīpa, despite such ambivalence, persisted and became a part of the Brahmanical concept of the universe, being sometimes identical with Bhāratavarșa, and sometimes Bhāratavarşa being a part of it. That the term was used as a reference point for actual geographical space is seen in Mauryan emperor Aśoka's reference, made in the third century BCE, to Jambudvīpa as the space where 'the gods, who were formerly unmingled with men, have now become mingled with them'. Aśoka's Jambudvīpa, over which he ruled, if taken to correspond to actual geographical space, extended from Afghanistan to the Deccan, including areas outside the Indian subcontinent and excluding south of the subcontinent. Jambudvīpa, part of an elaborate cosmography, in which the earth consisted of dvīpas or islands, was also a concept of a real country in the sense that familiar janapada names and places could be located within it. Bhāratavarșa too was a crucial part of an elaborate cosmographic schema, but as will be clear as we present an outline of the schema, it could also be taken to correspond to a geographical space or a framework within which, over time, different constituent regions could be located. In the early stage of its use, it seems that the term Bhāratavarșa did not carry the meaning which it came to be associated with later when it could correspond vaguely with the geographical limits of the Indian subcontinent. The historical stages of the expansion of that meaning are not, however, clear. Khāravela, the king of Kalinga or coastal Orissa in the first century вс, claimed, in his epigraph, to have gone out to conquer Bharadavasa (Bhāratavarşa) in his tenth regnal year, it being one of many such expeditions that he undertook. Clearly, Kalinga was not seen as a part of Bhāratavarsa when he was ruling. In fact, Bhāratavarşa figured as a key component in an elaborate cosmographical schema only when the Purāṇas were being compiled, and it is to this kind of textual evidence that we have to turn to understand the structure of the space and its associated characteristics that the term conveyed. The cosmographic schema, of which Bhāratavarșa was a part, is available in more or less similar forms in a number of Purāņas, despite some variations in them, and for the first time, similar to the dig-vijaya narratives, as in the Mahābhārata, one comes across in them what emerged as a fully developed idea of Bhāratavarșa and its different spatial segments. The Puranic texts are voluminous, and it would also be pointless to attempt fresh comparison of material contained in them by taking the Purāņas individually. I would therefore limit myself to referring to the already much-used text of the Vişnu-Purāṇa. For purposes of comparison to ascertain both the consistency and elaboration of the concept in relation to actual geographical space, I shall turn to two other texts: the Raghuvamśam of Kālidāsa, particularly the part on the dig-vijaya of Raghu, and the tenth-century text Kāvyamīmāmsā by Rājaśekhara. The texts may be taken to represent a sufficiently wide span of time to illustrate not only how the Puranic schema and its details had become more or less stereotyped, but also how the meaning of the same details may have undergone some change. ## II. Bhāratavarșa in the Purāṇas A few preliminary points regarding the nature of Puranic evidence may be made before we turn to the material in the Vişnu-Purāṇa. The Puranic details, despite their characterization as 'geographical details', do not pertain to the geography of India. One section of the relevant part of the Purāņa is devoted to what is called Bhāratavarsa-varnanam, but that too within a broad design of the world as a part of the cosmos, interspersed with the story of creation, detailed genealogies to show the essential connection between genealogy and space, enumeration of broad divisions of the world and the location of Bhāratavarsa in it, as also enumeration of all the divisions within Bhāratavarșa and so on. Bhāratavarsa, before it can be considered as a geographical category, therefore needs to be taken within the entire context of its particular location, without which exploring the material only for premeditated selection of geographical names will obfuscate various possible meanings of the term. The details in different Purāņas vary, and there are internal contradictions within individual Purāņas. Not all Purāņic schemas are mutually reconcilable either. Nevertheless, it is the Puranas, apart from the epic Mahābhārata, which present us, for the first time, with what was perceived as the structure of Bhāratavarṣa; it is thus definitely worthwhile persuing the Puranic evidence for one looking at the concept of Bhāratavarșa and its significance for Indian history. In the Vişnu-Purāņa, Bhāratavarșa figures in the prathama-adhyāya (first chapter) of Book 2, titled Jagat-srsti-sambaddha-Bhārata-vamśa-kathanam (narration of Bharata genealogy connected with the creation of the universe). Bharata-vamsa here is interchangeable with Svayambhū-vamśa because the lineage starts with Svayambhū Manu. In this genealogy, seven sons of Manu were put in charge of seven dvīpas or islands (Jambu, Plakṣa, Śālmalī, Krauñca, Kuśa, Śāka, Puşkara) which together constituted the earth (Vasundharā). Viṣṇu-Purāṇa's Bharata-vamśa-kathanam is followed by Jambudvīpa-varņanam (description of Jambudvipa), then Bhārata-varsa-varnanam (description of Bhāratavarsa), in turn followed by sad-dvīpa-varnanam (description of six dvīpas). Jambudvipa, of which the ruler was Priyavrata's son Agnidhra, was, in turn, divided into nine parts. Of those, Himavarsa, later mentioned as Bhārata-varsa came to be ruled by Nābhi. Bharata, son of Rsabha, belonged to this lineage, and 'the country was termed Bhārata from the time it was relinquished to Bharata by his father...' The genealogy continued after Bharata, and Bhāratavarşa came to be divided into nine portions (bhedāh). “This was the creation of Svayambhūva Manu, by which the earth was peopled, when he presided over the first Manvantara. ' Cosmographic details continue in the second and third adhyayās of Book 2 of the Vişnu-Purāṇa, with the third adhyāya beginning with the following verse: > Uttaram yat samudrasya Himadreścaiva dakșiņam varşam tad Bhāratam nāma Bhārati yatra santatih A literal translation of the verse would be: > That [varsa] which lies to the north of the ocean and to the south of the snowy mountain, is called Bhārata, where the progeny is called Bhārati. In considering the possible meaning of the geographical and cultural space indicated in the verse quoted above, it is necessary to remember that it is located within a context which is not strictly, and correctly, geographical but cosmographical, although different natural landmarks, such as mountains and rivers, associated with different mountain ranges mark Bhāratavarșa out from other varsas and dvīpas, and imbue it with a geographical meaning. The mountain ranges, each a Kula-parvata (family mountain), are Mahendra, Malaya, Sahya, Śuktimat, Rkşa, Vindhya and Pāripātra, and rivers originating from them flow in different directions. It is the directions again, in combination with Madhyabhāga, or Madhyadeśa, which constitute the structure of Bhāratavarșa. Thus, although Bhāratavarșa is mentioned as being divided into nine divisions, all of which are again specified by their individual names, it is dik, or direction, which indicates how different communities of different janapadas were located in Bhāratavarșa. To cite a verse from the Vişnu-Purāṇa: > On the east (Pūrve) of Bharata dwell the Kiratas... on the west (Paścime) Yavanas; in the centre (madhye bhāgaśaḥ) reside Brāhmaņas, Kshatriyas, Vaiśyas, and Śūdras, occupied in their respective duties of sacrifice, arms, trade, and service. Dik, or direction, is here used to suggest the centrality of the middle zone not simply in the geographical sense, but to suggest a contrast between what the Vişnu-Purāṇa, as indeed other Purāņas too, considered to be a model social order, distinct from the order prevalent in outlying areas in the east and on the west. The same centrality of the middle zone is carried over in the context in which janapada communities are sought to be geographically located. Thus, the Kurus and the Pāñcālas are assigned to the middle region (Madhyadeśa); the Kāmarūpas to the east (pūrva-deśādika); Sauraştras and Ābhīras to Aparānta (west), and so on. From the way Bhāratavarșa is represented in four consecutive adhyāyas of part two of the Vişnu-Purāņa (or as it has been represented, sometimes in greater detail, in other major Purāņas) as a part of a cosmographical package, several points regarding the nature of this representation seem to emerge. One, since Bhāratavarșa is very clearly a component of a much larger design, methodologically it may be inappropriate to identify the component with a concrete territorial unit and take it to represent a geographical reality. At the same time, in the construction of the structure of Bhāratavarsa, the pool of current geographical knowledge as well as the understanding of the cultural attributes of the structure were put into use by the compilers of the Puranas. Thus in designing Bhāratavarșa, the basic cartographic principle of dividing up space, first into four cardinal directions with a central zone in the middle, and then making further divisions of seven or nine, could be followed for locating its inhabitants in their respective zones, radiating to the east, west, north and south. This geographical knowledge of the early texts does not always correspond to what is known from various other sources and from other contemporary information about locations of various ethnic communities. Thus, in the Visnu-Purāņa itself, the Pundras, Kalingas, and Magadhas are all clubbed together with the southerners or the Dāksiņātyās, and the list, purporting to be that for the western region (aparāntāh) includes, without any references to the northern direction, a number of communities such Sakalavāsin, Sālva, Madra, Hūna, Saindhava and so on, who would otherwise be located in the northern region. Before thus taking the description of Bhāratavarsa in the early texts to correspond to the geography of our Indian history, it is necessary to note that to the compilers of these texts it was perhaps not the accuracy of detail, but the overall structure of Bhāratavarsa and the way it fitted into a cosmographic design which were more relevant. Second, the description of Bhāratavarsa in the Purāṇas is inextricably linked with details of genealogy which envelops different layers within the design of the universe, originating with the sons of Manu and with the original division of the patrimony among seven sons. In the description of Bhāratavarsa in the Vişnu-Purāna, the expression Bhārati Santatih has therefore to be taken not in the exaggerated sense of the 'children of Bhārata', but simply as 'children born in the lineage of Bharata'. This sense is made clear in a parallel expression which occurs in the Vāyu-Purāṇa. > Tair = idam Bhāratam varsam nava-bhāgair-alamkṛtam Teşām vamśa-prasūtais = ca bhukteyam Bhārati purā This Bhāratavarsa, adorned with its nine parts, was enjoyed in the past by those born in their family and known as Bhāratī. Third, that the meaning of Bharatavarșa went beyond a geographical sense is conveyed in the way it was distinguished from other varsas of Jambudvipa. Thus, while in other varșas there was no calamity, no fear either of growing old or of death, no sense of dharma or adharma, or of the high and low, or of the division of yugas, Bhāratavarșa alone journeyed through various yugas; it was the region where Karma was in operation and which was bhoga-bhūmi, but by virtue of its being karma-bhūmi, Bhāratavarṣa was projected as the best among all other varșas. The other varșas were bhoga-bhūmi, but by virtue of its being karma-bhūmi, Bhāratavarṣa was projected as the best among all other varșas. Since the Puranic projection of varsa is inextricably linked with genealogy of rulers, another dimension of space, conceived as the domain of a sovereign, or of one aspiring to be a sovereign, may be examined by referring to the concept of dig-vijaya or the 'conquest of quarters'. We set out to explore now how this can lend further insight to our understanding of the meaning of Bhāratavar. sa, by referring to the details of Ikṣvāku ruler Raghu's dig-jigīṣā or 'the intent of conquering the quarters', specifically portrayed in Kālidāsa's genealogical poem Raghuvamśam. ## III. Conquest of the Quarters (Dig-jigīsā) Raghu, in the way Kālidāsa described his mission of conquests, started with the east (prācī), it being in order to mention the direction first, and this brought him to the shore of the ocean, dark with groves of palm trees. In this region he encountered the Suhmas, the Vanga princes with their fleet, and, then, having crossed the river Kapiśā, he proceeded to Utkala. Moving further, beyond the summit of the Mahendra mountain, he subdued the Kalingas. Along the seashore then, the army marched in the direction taken by sage Agastya (i.e. south), taking them to the river Kāverī and to the valley of the Malaya mountain. The other landmarks mentioned in the region are the confluence of the river Tāmraparņi and the ocean, and the Malaya and Dardura mountains, which were like 'the breasts of that region' (diśastasyāh). From there, marching further, Raghu crossed the Sahya mountain. The subjugation of Aparānta and Kerala were followed by Raghu's thrust toward Trikūța from where the move was in the direction of the Parāsikas by a land route (sthala-vartmanā) and, in the same context, mention is also made of the Yavanas. Raghu's fierce encounter with the western peoples, adept as cavalry men (pāscātaih aśva-sādhanaiḥ), resulted in the following: > He strewed the earth with their bearded heads severed (from their bodies) by his (Bhalla) arrows as with honeycombs covered with swarms of flies. The remnant, removing their helmets, threw themselves upon his protection. Raghu's march to the north (Udicya), the quarter presided over by Kubera, brought against him the Hunas and the Kambojas. Journeying in the north further brought him to the lofty Himala-yan ranges and to the upper reaches of the river Ganga, where he encountered the Kirātas and other mountain tribes (pārvatīya gana). Beyond Mount Kailāsa and having crossed the river Lauhi-tya, Raghu reached the kingdoms of Prāgjyotișa and Kāmarūpa, and with their subjugation was completed Raghu's conquest of the quarters, a prelude to the performance of Viśvajit ('conquest of the world') sacrifice. With the conquest of dis and the performance of the sacrifice, Raghu's accomplishment of the status of a sovereign ruler, which is the ultimate aspiration of all vigīsu rulers, was completed. The major relevance of the Raghuvamsam material that we have attempted to summarise is the convergence of the geography of space defined by the four dis and the space over which an early Indian monarch aspired to have unrivalled dominance, the idea being conveyed also by another expression, cakravarti-kşetra. However, from the purely 'geographical' point of view what strikes us as interesting are references to different communities, natural landmarks and to the fauna and flora, which are pointedly connected with each dis. For the sake of convenience, the material may be presented in a tabular form. These references in the Raghuvamsam, though not purporting to give us the structure of Bharatavarșa, nevertheless use the same key concept of dis or quarter. Kālidāsa's idea was not to provide strict geographical accuracy, but to use the schema for delineating the space over which a sovereign-designate had to traverse. One can easily point out the arbitrariness of some locations: the southern region being suggested only after Kalinga; the mention of Kerala only after Sahya or the Western Ghats and not in association with Malaya; the separation of Kāmarūpa from Prāgjyotișa and their implied location in the context of Raghu's expansive expedition in Udicya. It may be reiterated that Kālidāsa appears more concerned with making Raghu's itinerary conform to the key concept of four quarters than with geographical accuracy. In fact, by using the concept of dis, starting with the east, and, at the same time, ending this itinerary with Prağjyotișa and Kāmarūpa (which too should properly be located in prācya), Kālidāsa appears to have curiously combined the concept of four directions with a circular journey encircling the space of a sovereign ruler. The Raghuvamśam may also be said to have adequately used the existing geographical knowledge of four cardinal directions, of major physical landmarks (the oceans, the mountains, the rivers) in relation to these directions and of the locations of different communities in relation to them and physical landmarks. The additional significance of the Raghu-vamśam's description is that it shows a degree of familiarity with distinctive types of flora, fauna and other products of different locations and with what are projected as special characteristics associated with different ethnic groups. Some of the significant features of Kālidāsa's material will be highlighted again when we sum up our findings on the concept of Bhāratavarsa. We may now turn to a later text, composed in the ninth-tenth century, which has a completely different kind if concern, but which nevertheless incorporates, and in a sense, reiterates, the earlier notion of Bhārata-varșa in its relationship with cakravarti-kşetra, i.e. 'the field of one who moves on uninterrupted'. ## IV. An Exercise in Synthesis? Rājaśekhara's Kāvyamīmāmsā The idea of Bhāratavarsa, as articulated in the Purāņas, as consisting of various communities and janapadas, and continuing, without using the term Bharatavarsa in Kālidāsa's fascinating though brief account of the conquest of the four quarters, gradually became more or less stereotyped, but with modifications and elaborations in individual texts. The Kāvyamīmāmsā ('Discourse on Poetry') of Rājaśekhara arrives at the theme of deśa-vibhāga (division of deśa, i.e. 'country') after a thorough discussion of what the terms Jagat (universe) and Bhuvana (world) mean, since: (i) the entire Jagat or Bhuvana, or only a part of it may mean deśa, and (ii) in dealing with deśa and kāla (time) the poet is required not to display poverty of understanding (arthadaridratā). It is in this context of the discourse of deśa that Rājaśekhara locates Bhāratavar. sa after asserting that given the possibility of the existence of many Bhuvanas (worlds), bhūloka means the earth (Prthvi) which consists of seven islands. Despite resorting to the practice of citing many opinions, Rājaśekhara essentially follows the Purāņic structure of Bhāratavarşa, particularly the one elaborated in the Vayu-Purāṇa, and although he too divided Bhāratavarșa into nine parts, he sticks to the convention of arranging the various janapadas or localities associated with individual communities in terms of quarters. What strikes as significantly innovative in Rājaśekhara's design is that he suggests specific geographical points from which different quarters begin, his central or the core region being the well-defined Āryāvarta, equivalent to Madhyadeśa, and geographically defined as the space between the eastern and western oceans and between the Himalayas and the Vindhyas. In Rājaśekhara's attempt to specify the beginning of each quarter (dis) in relation to Aryavarta, it becomes obvious that overlaps are inevitable, more so when Rājaśekhara suggests that divisions of dik, despite being unstable to some thinkers, could be made from antarvedi with Kānyakubja (Kanauj) being something like a meridian. With both Āryāvarta and Antarvedi (region between the Yamuna and Gangā) being considered the centre for defining the cardinal directions, it is understandable that all directions overlap noticeably with the centre. Thus, although Āryāvarta is defined as the region between the eastern and the western sea and between the Himalayas and the Vindhyas, the Pūrvadeśa (eastern region) starts, according to Rājaśekhara, from the east of Vārāņasī, the western region (paścād-deśa) from Devasabha, the southern region (Daksiņāpatha) from Māhīṣmatī and northern region from Prthūdaka. This anomaly notwithstanding, Rājaśekhara's attempt describe the details of the structure of a quarter (dis, dik) in relation to an actual geographical point (such as Vārāņasī, Māhīṣmatī, Prthūdaka, etc.) suggests a recognition and use of such points as important enough to define the beginning of each quarter and thus relate them to the constituent elements of the quarter. In Rājaśekhara's detailed treatment of the structures of individual quarters, one can notice the presence of three main Purāņic elements: the enumeration of the janapadas, the mountain ranges associated with the quarters, and the rivers flowing from them. This is a pattern which, with expected variations, is present in all s. However, what Rājaśekhara does additionally is to add, in the fashion of Kalidasa, the list of natural products which can be associated with a quarter. To give a single example, Uttarāpatha (