Sinhala group emotions and the Sanskrit cosmopolis

[Full title: A blot on the landscape? Sinhala group emotions and the Sanskrit cosmopolis]

Sheldon Pollock’s analysis of a millennium (c. 500–1500) of South and Southeast Asian history, in his The Language of the Gods in the World of Men, adopts the kind of genuinely panoramic comparative sensibility of world history at its best. Theoretical reflection on identity formation has usually focused on the modern period in which the influence of Western power and ideas bulks large; there have been few serious attempts to compare how diff erent cultural regions may have construed group identity in the premodern world. Pollock’s approach is particularly notable for the attempt to avoid the redeployment of concepts lifted semi-consciously from European history. It is only when a work such as this appears that we realize how much scholarship has lacked the conceptual rigour provided by the comparative method.

Pollock describes the Indic world in terms of a Sanskrit cosmopolis (a ‘transregional culture-power sphere of Sanskrit’) and then shows how, from the turn of the millennium, in region after region, this gave way to the development of vernacular literature.^[1] These new literary cultures were deeply influenced by or even modelled on the Sanskrit inheritance, but they were now turned inwards to address a more local audience in its own language.^[2] Pollock draws a striking parallel with the chronology of an equivalent process in Europe’s turn from Latin to the vernacular. However, Pollock argues that vernacularization did not carry the same political and social implications in the Indic world: it did not represent the first steps in a journey towards nations, or even ethnic groups, but was rather the occasion for the development of ‘vernacular polities’.^[3]

It should be noted at the outset that there is some ambiguity in The Language of the Gods as to how far Sri Lanka is to be understood as a participant in the dynamics that shape the trajectory of the Sanskrit cosmopolis. Pollock recognizes that Sinhala has its deepest origins as a written language as far back as the centuries bce,^[4] and more importantly that Sri Lanka also participated in the Pāli ecumene, such that ‘Sanskrit may never have occupied a dominant posi tion in either aesthetic or political expression’.^[5] We shall return to this point. Nevertheless, Pollock notes that the transformations in Sinhala literary culture from the end of the first millennium bear a strong resemblance to developments elsewhere. As a result, he refers to these developments a number of times, where they are depicted as an unusually early case of his central thesis–particularly exemplified in the ninth-century Siyabaslakara.^[6] Moreover, presumably Pollock’s many generalizations about the functioning of identity in ‘South Asia’ or the cos-mopolis more widely are intended to include Sri Lanka.

One such insight is the way that the vernacular polity produces a distinctly bounded concept of its proper arena, so that we see in Lanka for the first time in the tenth and more strongly the twelfth centuries, ‘the earliest literary represen-tations… of a newly coherent geocultural space’.^[7] The language of universal sovereignty appropriate for the old cosmopolis may still linger on in a symbolic form (so that a king is hailed as cakravarti with Jambudīpa-wide hegemony), but this is understood to refer in actuality to an identified territory.^[8] Java presents an inter esting comparison with Sri Lanka in this regard.^[9] We need not thereby underplay the extent to which Lankan kings looked to extend their influence into South India and even Southeast Asia whenever they were strong enough. Yet Lankan kings rarely were strong enough to launch and sustain such projects; they never sought lasting dominion overseas.^[10]

While this may indeed be comparable with developments in Europe, it is also true that in other ways the Lankan political imagination seems firmly branded with a certain ‘cosmopolitan’ quality found in societies across the wider Indic world. In terms of its religious self-f ashioning, Lankan kingship could aff ord to be much more eclectic and flexible than its Christian counterparts, borrowing extensively from diff erent Buddhist traditions as well as those now considered ‘Hindu’. The latter included notions of divine kingship that rendered the king ‘as a type of axis mundi cum avatar’.^[11] Brahmans–as indeed across the Indic world–played the primary role in conducting the ritual life of the court. In this way, Sinhala kings could speak to their dynastic rivals across the seas and co-opt their claims to sacralized authority.

Indeed, if sovereignty was considered as geographically confined, it was not therefore culturally circumscribed, for the island itself was home to diverse groups. A claim to rule over Lanka entailed dominion over the Tamil-speaking parts of the north and the east, over the väddā of the interior, and over the mercantile and fishing groups of diverse origins strung along the coasts, as well as over the Sinhala-speaking population. Nor was such diversity simply an uncomfortable fact, but rather something that flattered the authority of the king or cakravarti. To that extent, a cosmopolitan or ‘universalist’ notion of kingship remained. In keeping with this general celebration of pluriformity, mastery over several languages remained an ideal for scholars–indeed for kings–well into the vernacular period and beyond.^[12] By the time that the Portuguese arrived then in the first years of the sixteenth century, they found an elite culture that looked anything but monolithic. The kings of Kōṭṭe were dependent on quite recently arrived mercenaries from South India, and tried to tempt Karāvas to settle with land grants.^[13] Bhuvanekabāhu VII (1521–51), the last king or ‘emperor’ of Kōṭṭe to resist Portuguese control, signed his letters with a phrase in Tamil. Nor was the ‘digestion of the foreign’ always a bargain premised on cultural assimilation: hence Muslims could be incorporated into the Kandyan socio-political order in a way that was both profound and tolerant of their faith.^[14]

It is always salutary to remind ourselves of these facts, and careful analysis with Europe may indeed reveal some of these features to be rather distinctive of a South Asian or even Indic sensibility. Other features, however, may turn out to be applicable to premodern politics more generally, only acquiring any contrastive analytical relevance when set against some more modern visions of the nation-state.^[15] For example, Britain in the period 1110–1400 ce would present an analogous example of an island populated by a number of distinct peoples accord-ing to their own self-understandings, and yet according to R. R. Davies, ‘it was considered a sign of political maturity and power that a ruler governed several peoples rather than one’, with rulers from Athelstan (r. 925–39) onwards grant-ing themselves grandiloquent imperial titles as a result.^[16]

But what might all this mean for the possibility of ethnicity or even politi-cized ethnicity?^[17] Pollock’s stance here is radically different to that of the modernization theorists for whom the basic conditions of life in the premodern period everywhere rendered national sentiments or even politicized ethnicity impossible or superficial. Rather than being held to be dependent on a certain stage in material and political development, Pollock holds the existence of such sentiments to depend essentially on a cultural matrix of remarkable longevity. Hence they may be a foundational element of European life from very early on and yet irrelevant to the Indic world. For, as he sees it, ‘ethnic fictions’ or origin stories simply found no foothold in the Sanskrit imagination.^[18] Genealogy may have been a crucial element of dynastic aggrandizement, but we find no explicit discourses on the origins of peoples, they did not become the subject of chronicles as they did in Europe, and ‘one is hard-p ressed to identify a single instance of the propagation of shared group memories’.^[19]

Of the pre-modern Indian texts that have been read,

Nowhere is it possible to point to a discourse that links language, identity, and polity; in other words, nowhere does ethnicity–which for purposes of discussion we may define as the political salience of kin group sentiment–find even faint expression. It is equally impossible to locate evidence in South Asia for the linkage of blood and tongue so common in medieval Europe, or for cultures as associations restricted by so-called primordial ties.^[20]

This definition of ethnicity places a lot of weight on ‘fictive kinship’, and in one sense this is productive for comparative purposes.^[21] It serves to highlight what may well be a significant breach between the Indic and European. According to Davies, myths of biological descent were ‘the veriest commonplaces of the histor-ical mythology of the medieval world’.^[22] In an emic sense, peoples were commu nities of common descent.^[23] For John Watts, by the fourteenth century in Europe more generally, ‘the hallmarks of nationhood were conventionally the ties of blood and language, though a common name, shared laws and customs, govern-ment and territory… were also often important’.^[24] In the surviving source material from Sri Lanka, on the other hand, the dynasty greatly overshadows any role taken by ‘the people’, and the question of the blood descent of the linguistic group had much less salience.^[25] In general, the sandeśa poems, and even the haṭanas (for example, the Sītāvaka Haṭana) seem more concerned with exalting royal lin eages and the island itself rather than the Sinhala people in an explicit sense.^[26] It is reasonable to suppose that the caste system, with its foundation in endogamy, presented an obstacle to larger group claims based on fictive kinship across South Asia–even if caste worked quite diff erently in Sri Lanka to many parts of India.^[27]

In other ways, however, the definition obstructs comparative insight.^[28] ‘Peoplehood’ may be imagined in a variety of ways: to make explicit claims about common descent a necessary feature of it is to foreclose too quickly on broadly comparable dynamics of group sentiment. For example, however much descent claims may feature in emic understandings of ethnicity in medieval Britain, Davies does not ultimately make it a necessary feature of his etic definition of peoplehood. Instead his definition is that of an imagined community–imagined by itself and/ or by others.^[29] This has much in common with Obeyesekere’s richer notion of the ‘ethnic moral community’.^[30] Two other points emerge from Davies’ analysis that will be helpful for us as we proceed: that the formation of ethnicity is not dependent on a single cohesive state structure; and that its subjective and often taken-for-granted qualities mean that it is not a phenomenon that implants itself in the historical record in a straightforward way.^[31] It is ‘alternatively a prominent and a strangely evanescent concept’.^[32]

Nevertheless, over the course of Pollock’s millennium, there is certainly evidence in the Lankan record for many of the features that are typically associated with ethnicity.^[33] Notwithstanding the suggestion that blood had a rather weak salience in expressions of Sinhalaness, it may be that the famous origin story for the Sinhalas was in fact taken as such a descent claim for most or all of its very long history. The story of Sīhabāhu and Vijaya–itself probably an amalgamation of various such origin stories–is present in its extended form in the Mahāvaṃsa of the late fifth or early sixth century ce. It describes the fortunes of the miscreant Vijaya, descended from a lion, who pitched up in Lanka and overcame the yakkhiṇī Kuveṇī.^[34] It was retold many times in both Pāli and Sinhala texts. It has, to be sure, been a source of much debate as to whether the Sinhalas of the Mahāvaṃsa narrative can be understood as a dynasty or a people. Yet, the likelihood is that these stories were already taken as describing the origin of a far broader group than a dynasty at the time the Mahāvaṃsa was composed. Vijaya arrives on the island, with 700 warriors, to find it uninhabited by humans but only by yakkhas: this is clearly a colonization myth.^[35] Rather intriguingly, the logic deployed here fol-lows that of many other ‘stranger-king’ stories across the world, which are usually taken to define the origin of both kingship and society tout court.^[36]

The question as to whether and when Sinhalaness went from being an elite/ d ynastic category to one that encompassed all Sinhala-speakers stimulated an academic controversy in the 1980s and 1990s between K. N. O. Dharmadasa (who argued for an early emergence, by the tenth century at least) and R. A. L. H.

Gunawardana (who argued for an initial emergence in the twelfth century but remaining muted until the nineteenth century). For our purposes, however, it is significant that both argue for an initial crystallization either before or within Pollock’s vernacularization period.^[37] One text that became the focus of debate is dated to the tenth century, the Dhampiyā aṭuvā gäṭapadaya, which was apparently written by the king Kassapa V (914–23)–and therefore has as great a claim as any to represent the ‘public transcript’ of communal origination.^[38]

Gunawardana gives the following translation of a critical excerpt (in which ‘helu’ is used as a synonym for Sinhala):

How is (the term) in the heḷu language derived? That is derived on account of residents in the island being heḷu. How is it that (they) are (called) heḷa? Having killed a lion, King Sīhabāhu was called Sīhaḷa (as in the Pāli phrase), ‘Sīhaḷa, on account of having cut (or killed) a (or the) lion.’ On account of being his progeny (ohu daru bävin), Prince Vida (Vijaya) was called Sīhaḷa. Others came to be called Sīhaḷa on account of being their retinue (ehuvu pirivara bävin).^[39]

Gunawardana notes that this is an inelegant translation because the exact meanings matter for his debate with Dharmadasa. Yet for all that Gunawardana wanted to push forward the salience of ethnic identity as late as possible, it seems clear that a significant group of residents of the island were known as Sinhalas by this time. Gunawardana argued that this group pertained to the ‘kulīna’ or ruling stra-tum, and that the lower classes were excluded from this group–but this element of the argument is not particularly compelling.^[40]

Either way, this does not sit easily with Pollock’s characterization of an Indic system in which ‘the names for the vernaculars seem to abstract them from the domain of the group and locate them in what almost seem ecospheres… we must conclude it was far more often region that made a language (and a people) than the reverse’.^[41] For what do we have in the Dhampiyā aṭuvā gäṭapadaya right at the beginnings of the vernacular turn? Yes, the name of the language derives from the land, but that in turn derives from residents of the land, who had taken their name from the lion-natured founding figures Sīhabāhu and Vijaya… The flow of causality in the development of identity described by this and later texts such as the twelfth-century Dharmapradīpikā are scholarly reflections and may or may not reflect what happened historically.^[42] The more significant point is that by this time, the land, heroic founding figures, the residents of the land and the language were all profoundly identified with each other and grouped under the same name.^[43] And that it was considered to be a matter of some significance to the highest elites how the name for the wider Sinhala group evolved and could be defined. In fact, a lineage-b ased understanding may have been current much ear lier than this (perhaps a century or so after the composition of the Mahāvaṃsa) to judge from the version of the Sīhabāhu story relayed by the Chinese traveller Xuanzang, which he must have heard from Lankan monks who had come to India.^[44] It says that after the founder arrived to the island and tricked his way to power, he propagated and reproduced, and because his off spring became numerous they divided into classes and elected a king and ministers. The men of the ‘Lion State’ (shizi guo) are described in racial terms, possessing certain physical and psychological characteristics as a result of their ancestry from a lion. This may well reflect a strand of Chinese racial ethnology, but Xuanzang, at least, understood the Sinhala origin story as a descent claim–and for a whole society (note that all classes are held to be descended from the lion-fathered founder).^[45] Indeed, there is surely an opportunity for drawing more on foreign deployments of the term ‘Sinhala’ from this period into our analysis.^[46]

Furthermore, even Gunawardana saw the period of Cōḷa occupation in the tenth to eleventh centuries as providing the conditions for an ‘archaic ethnicity’ (that is, one sensed by sections of the elite and literati). In his view, the crucial shift to an inclusive definition of ‘Sinhala’ had happened by the reign of Parākramabāhu I’s successor, Niśśankamalla (1187–96). The Cōḷas certainly provided the Sinhala kingdoms with more than enough alterity through which to fashion a contrasting image of themselves.^[47] Sinhala texts of the thirteenth cen tury, such as the Pūjāvaliya, show the lasting impact of these conflicts in inflaming identity politics. But the truth is that these texts were merely redeploying a discourse of alterity already formed in the Pāli vaṃsas.^[48] This sense of opposition to a ‘demaḷa’ foe has become a shared understanding, expressed and reinforced through more particular shared memories of particular wars and battles, as regis-tered in a range of texts in diff erent languages with diff erent purposes.^[49] It is worth pausing here to underline that of course Tamils or demaḷas are not only demonized in the Pāli and Sinhala texts; they are present in them in a number of ways. This is to be expected: near ‘outsiders’ usually play several roles in the reproduction of domestic society; they are allies, marriage partners, rit-ual specialists and more.^[50] This multivalency of the foreign is a theme that runs through Gananath Obeyesekere’s work, and is also pursued in Marshall Sahlins’ recent theories of the stranger-king: it is a key theme of this volume too.^[51]

When the Portuguese and Dutch began to produce more ethnological material about Sri Lanka in the seventeenth century, what they report is not at all at odds with what we have gathered from these clues in the Sinhala and Pāli texts.^[52] The European writers are quite clear that the Sīhabāhu-Vijaya narrative is an origin story for the whole people. Interestingly, this is not the only origin story they relate (‘and as all nations seek to glorify their origin, and the heathen almost always imagine fabulous ones’), which indicates both that the vaṃsa narrative was not the only source of self-understanding and that there was a continuing need on the part of the Sinhalas and their neighbours to create stories to explain where they and their name came from.^[53] Much more could be said in this vein–reflecting on the profound historical consciousness of the Sinhala texts for example–but we must move on.^[54]

Charles Hallisey has suggested that:

The very period in which we see Sinhala fully realized as a literary language–that is, around the turn of the millennium–was also the time that use of the term ‘Sinhala’ was extended… to the general population and their language.^[55]

If so, it would make sense to see this in the context of the first major unification of the island under the Okkākas in the tenth century.^[56] Land, language, people and polity all seem to be subject to a project of conceptual solidification and reorganization. What about religion? Again, in Pollock’s analysis this realm of life is aff orded only a strangely attenuated connection to both political and aesthetic projects. Again, for Sri Lanka, this makes little sense.

That a certain religious cosmopolitanism should emerge as a major charac-teristic of the Indic world–and one that distinguishes it in enduring and profound ways from Europe–will be a surprise to no one. Ultimately its roots must lie not in a particular quality of Sanskrit or its literary development, but in the basic nature of the ideas developed in the formative period for South Asian religious and philosophical thought. Here the salvationist or transcendentalist urge was not initially expressed in a monotheistic devotion to a jealous god, and a given path to salvation could co-exist with all manner of projects of truth-seeking or attempts to access divine power for earthly purposes. Pollock goes further, however, to argue that there was nothing in India comparable to the Providential charter some poets and stoic thinkers provided for the Roman Empire, and he presumably has in mind the ubiquitous and long-l asting theme of Christian Providentialism too.^[57] In India, by contrast the supreme deity was simply ‘irrelevant as a source of royal authority. A talismanic or apotropaic force? Yes. But a granter of heavenly mandate, a justi-fier of rule, a transcendent real-estate agent awarding parcels of land? Never’.^[58] No one would claim that Sri Lanka presented an exact analogy to these late pagan and Christian ideas in the West. For one thing, the language of monotheistic agency is quite inappropriate in the Pāli Buddhist context. Insofar as ‘Providence’ implies the continuing interventions of an all-powerful deity, it has little relevance to Lanka. Moreover, it has long been a concern for scholars of Sri Lanka to object to modern nationalist-Buddhist readings of the past that have often assumed the timeless truth of very similar ideas. Yet the main dimension of Providentialism at issue here is the provision of a sacred mandate, and in this sense it does indeed become possible to discern an analogous logic as far back as the Mahāvaṃsa. Who can doubt that it is a major task of the Mahāvaṃsa to celebrate the special relationship between Buddha and the blessed realm of Lanka, to establish how deeply the Buddha impressed his dhamma on the land by his three visits, just as he physically sank his foot into the mountain of Sumanakūṭa (known today as Adam’s Peak)?^[59] Or that the kings of Lanka are thus charged with a cherished role in protecting his teachings against its enemies, which will in turn mean that they must flourish?^[60]

We need not rehearse here well-known features of the Mahāvaṃsa’s account of Duṭugämaṇu’s war against the demaḷas from South India, so bluntly framed as a sacred war.^[61] As just one detail, consider the story of one of Duṭugämaṇu’s ten preternaturally gifted warriors. Nandhimitta had grown up in Anurādhapura and witnessed the demalas of Elara’s occupying force desecrating the thüpas, so he went among them tearing them apart and throwing their body parts over the walls.^[62] Then he reasoned that such negative attrition against the enemy was not as good as positively serving the religion; if he waged such war on behalf of the (Sinhala) kings of the south then he could add lustre to the faith. When the post-vernacular Thūpavaṃsa of the thirteenth century relays this episode, it need only elaborate the same theme by adding a few pungent details–that the demaḷas had been defecating and urinating in the courtyard of the sacred Bodh tree, for example.^[63] The line of alterity against the demaḷas from the vaṃsas to the most demotic war poems of the early modern period, is thus thickened with the senti-ments of religious opposition.^[64] To clarify: it is not that Pollock discounts the possibility of the mobilization of religious alterity elsewhere in South Asia. Indeed, in other publications he has made such a claim with regard to the Rāmāyaṇa’s role in the face of aggression by Muslim powers.^[65] It is rather that in the Lankan tradition a) this alterity may be expressed by a quasi-Providential connection between the fate of a religious project and the fate of a people and land, and b) was thus liable to become fused with other forms of group emotion.^[66]

Naturally, the assertion of religious superiority also had an autonomous dynamic: it was never merely the waging of ethnic or dynastic warfare by other means. The Mahāvaṃsa–giving voice to the Mahāvihāra tradition that would eventually come to reign supreme from the twelfth century–tells us of monks and kings concerned to put down heretical doctrines, by burning books if neces-sary. It is notable that the chief ‘heretical’ threat the Mahāvihāra identified during the reign of Vohārika Tissa (214–36 ce) came from the Vaitulyas, who seem to have taught something rather close to monotheism, or at least theism.^[67] But we know that elsewhere in the world such theological projects have often provided an irresistible ideological resource for political elites; nothing suggests Lanka was diff erent in that regard.

If the quasi-Providentialist theme begins well before the shift to vernacular literature, once that shift was underway it was elaborated in more explicit terms. In a tenth-century inscription we have the first assertion that ‘none but future Buddhas would become kings of prosperous Lanka’.^[68] By this time, South Indian alterity was being provided by the Pāṇḍyas and then Cōḷas.^[69] Now, Pollock does note in parentheses that the Cōḻa Empire, with its aggressive assertion of Saivism, stands as a rare exception to his remarks about the prevailing religious pluralism. By the ninth century, the Saiva canon was already shaping up against Jain and ‘Theravāda’ opponents.^[70] It would seem, then, that along this southern periphery of the Sanskrit cosmopolis we have the sacred plunged into the dirty work of political competition and group struggle in a way that is all too familiar from other parts of the world (but is unmatched elsewhere in Pollockian India). This generated a pugnacious rhetoric about the sanctified morality of warfare that one could easily find parallels for in Europe.

It is always worth emphasizing that ideological strategies of boundary-hardening may work by co-opting or mimicking the discourse of those the bound-ary is hardened against. The following quotation, taken from Charles Hallisey’s translation, is pertinent for many reasons.

It is an inscription in Sinhala by Niśśankamalla (r. 1187–96):

King Niśśanka Malla ensured the long stability of the state and the religion (Lokaśāsanaya). Moreover considering that the island of Lanka is a noble land because of the establishment of the sāsana there, that the living beings in it have lofty excellences and that, therefore, they should receive advice and protection, he out of compassion, proclaimed the following maxim of good counsel:

Though kings appear in human form, they are human divinities (naradēvatā) and must, therefore, be regarded as gods. The appearance of an impartial king should be welcomed as the appearance of the Buddha… If the wishes of kings were not observed, the human world would be like hell, but if their wishes were respected it would be like heaven.^[71]

Such a bald assertion of the divinity of the king is rare in a Lankan context and possibly due to a desire to assert the kind of authority that his Indian counterparts possessed: this is cosmopolitanism as a function of royal peer-to-peer relations. Yet the inscription also speaks of the creation of a moral community rooted in a space at once political and geographical, which has kingship at its core and the advancement of a specific tradition of religious teachings as its purpose. Hallisey must be right to connect this to Niśśankamalla’s own origin in Kālinga and his unclear claims to the throne.^[72] It is precisely because of the cosmopolitan origin of the king, and the cosmopolitan audience for his claims, that he must engage in a profound emphasis on localization. One would be hard-p ressed to avoid interpreting these assertions as an unusually blatant exercise in ‘legitimization’.

Nor can such immediate and contingent forms of legitimization simply be interpreted as evanescent strategies. Rather they streamed into and inflected a more enduring discourse that was picked up repeatedly in subsequent texts. The thirteenth-century Pūjāvaliya has been much-quoted for the way that it aggres sively asserts that only Buddhist kings protecting the sāsana should–and indeed would–be able to rule over Lanka.^[73] Any non-believers must simply fail just as the Buddha had banished the demons. This was written following the invasion of the Kalinga King Māgha (1215) who stimulated the ire of the vaṃsa-writers ever after with his book-burning brand of Saivism. The Cūḷavaṃsa accused him of forcing people to convert to ‘wrong views’: ‘Thus the Damila warriors in imi-tation of the warriors of Māra, destroyed in the evil of their nature the laity and the order.’^[74] In the midst of this ‘Tamil conflagration’, monks fled to a retreat in South India associated with a ‘Cōḷaganga’, and when Buddhism was restored under Parākramabāhu II (1236–70), mahathēras from India were brought over to assist.^[75] The sense then of the wider Buddhist ecumene was still present, but there is little doubt that Māgha’s unusually aggressive religious policy had stimulated another flash of electricity through the connection between religious community, sovereignty and land.^[76]

Pollock argues that the connection between popularizing religious movements and vernacularization has been greatly exaggerated, that the original impetus for the vernacular came from the secular aesthetic concerns of courts and only later did a few regional vernacular movements emerge with some sort of religious purpose.^[77] Whether this chronology fits Sri Lanka can be left to schol ars with more expertise in this area.^[78] But it may be artificial in the Lankan case to perceive two distinct vernacular flowerings, a cosmopolitan version divorced from religion and then a regional ‘counterrevolution’ closely linked with religious community. According to Jonathan Walters, the popularization of Buddhism can also be located in the tenth to twelfth centuries.^[79] Certainly, whether they constitute a ‘second wave’ or not, various thirteenth-to fifteenth-century Sinhala texts could not be more explicit that their function was to make Buddhist teachings widely available to the non-learned.^[80] The Sinhala Thūpavaṃsa, for example, reflects (or seeks to drive) the development of a cult of the relics in Lanka, and this is a major mechanism by which sacred power is concretized for a mass audience, or a population gathered into an imagined community of Buddhist devo-tees.^[81] (Indeed, we shall see the continuing success of this project in the crucial role played by the Tooth Relic in political conflicts till the last days of the Kandyan kingdom.) It is likely that the identification between Buddhism and Lanka was further strengthened by the disappearance of Buddhist institutions in India from around this date.^[82]

Notice that we have not yet broached modernity–or even what some scholars now refer to as ‘early modernity’. When we come to the latter end of the premodern, Pollock does explore the possibility that the more northerly parts of South Asia may have experienced a form of alterity in the shape of the encroaching dominion of Muslim elites, and that in a few cases this might possibly have stimulated a vernacular response and a religious repudiation.^[83] Ultimately, how ever, he is reluctant to assign this much causal weight.^[84] A little later, Lanka had its own marauding monotheists to contend with, and these off ered yet another manifestation of alterity. In general, no scholar has been willing to compare the Lankan experience of the introduction of European Christianity with the Indian experience of the introduction of West and Central Asian Islam further north. This is no doubt because the modern-day politics of the historiographies are so diff erently organized. Yet a range of practices–above all the controversial facts of temple-destruction–present points of comparison. Whatever current scholarship is prepared to concede for the northern case, it is easier to acknowledge that the aggressive exclusivism of Portuguese Catholicism stimulated intermittently more resistant and boundary-hardened forms of Buddhist sentiment.^[85]

Pollock successfully isolates state formation and royal ambition as the main context for Sanskritization, yet the connection between the adoption of the whole Sanskrit imaginaire and the political will of kings is left ultimately mysterious. Underlying this is an initially laudable desire not to reduce cultural production to a matter of political instrumentality.^[86] But so uncompromising is his theoretical antagonism to the notion of ‘legitimization’ that Indic culture ends up as divergent from the West even at the level of the most basic mechanism by which religion relates to politics. Hence ‘ideology’ may be a permissible concept for ana lysing Europe but not South Asia.^[87] Elites did not ‘need’ religion, apparently, or ‘use’ it or even ‘benefit’ from it politically: in the Sanskrit mode of being, they simply lived and dreamed within its embrace.

In the case of Lanka, however, it seems that the celebration and imaginative evocation of the geography of the island of Lanka, the assertion of the moral supremacy of Buddhism, the elaboration of Pāli and Sinhala as literary languages, and the coalescence of Sinhalaness as an ethnic identity may all be related to the struggle to maintain a form of political unity and independence from foreign powers.^[88]

It is not, after all, hugely surprising that Sri Lanka should sit uncomfortably within some of Pollock’s generalizations: it is the fate of ambitious attempts at model building to be set upon by specialists agitated by the righteous indignation of the counterexample. It may serve to highlight a general problem with the method, however, which is to insist too much on the causal priority of cultural diff erence. This amounts to the exaltation of a diff erent Indic ‘ontology’ from which various ogres of liberal academia have been gratifyingly banished: thus the Sanskrit world did not have ‘empire’ but cosmopolitics; it did not give rise to the nation-state but to ‘vernacular polities’; it did not have ‘race’ or ‘ethnicity’ but entirely diff erent kinds of groupings; and it was even free from ‘ideology’ itself insofar as its dealings with the sacred were simply sincere.

But let us assume, for the moment, that Pollock is right about the implications of vernacularization in the Sanskrit cosmopolis in general but Sri Lanka fails to fit the model in the ways addressed above. In that case, the more interest-ing question is why that should be so.^[89] Unfortunately, it is also more diffi cult to answer. In his expressed reservations about Sri Lanka, Pollock refers to its partici-pation in another cosmopolitan order, that of Pāli imaginaire. Indeed, the more that we privilege the foundational role of the Pāli worldview in Lanka, the more Lanka could even be seen as ‘the exception that proves the rule’ of Pollock’s thesis. Does it show what happens when part of South Asia, so obviously part of the Indic world in so many ways, comes under the sway of a diff erent literary or even cognitive regime? Steven Collins had already pointed out the diff erence between ‘the Indic model of clerical-ideological power’ in which Sanskrit was imposed as a medium of expression without seeking to impose a single belief system, and the Pāli Theravāda model in which there was ‘an isomorphism between a sin gle language and a unitary ideology’.^[90] But given that what is at issue here is not Buddhism per se, which embedded itself in Asian societies in myriad ways, but the way that it was localized in the resurrected Pāli texts of the Lankan vaṃsas, to appeal to Pāli is therefore only to rephrase the question about Lanka.^[91] If the ‘Pāli cultural package’ adopted by the Southeast Asian centres was first formed in Sri Lanka, we are pressed again to consider what may have made the history of iden-tity formation in the island distinctive.^[92]

Victor Lieberman’s recent work of comparative history may provide us with the beginnings of an explanation framed in geopolitical terms.^[93] He places great emphasis on variations in how ‘protected’ diff erent parts of Eurasia were. In the first instance, this refers to whether they were vulnerable to conquest by Inner Asian warrior groups, but we could extend it more broadly to long-term openness to all kinds of conquest elites. The island of Lanka, as the Sanskrit poets were fond of pointing out, is surrounded by the ocean. However clichéd its island status has become and however much the global history of today tends to emphasize connectivity–yet those island boundaries mattered profoundly.^[94] They presented a natural (although by no means implacable) limit to the political ambitions of Sinhala rulers from earliest times: there was really no need for any ‘contraction to the vernacular polity’ that Pollock discerns elsewhere. More significantly, the ocean meant that while Lanka was open to the extension of military and political might by South Indian powers from at least the mid-first millennium ce, these projects were never strong enough to sweep away rule by Sinhala-speaking Buddhist rulers for more than a few generations.^[95]

Pollock refers to the ‘veritable law of political entropy’ in India such that few ruling lineages lasted more than two to three centuries.^[96] Most significant for us is to note the relatively weak continuity of identity between the dynasties riding these cycles in the subcontinent, even if there were very long-running chains of imperial imitation and charter state referral. In Sri Lanka, the sense of the continuity between diff erent instantiations of Sinhala kingship sustained by its relationship with the sāsana was presumably more profound. This may be partly because of its long gestation in Anurādhapura, which had its origins in the mid-first millennium bce and was not abandoned until the very end of the first millennium ce and was the unrivalled centre of political life in the island for that time. It would be diffi cult to find an equivalent in India over the same vast stretch of time. After that, as Sri Lanka was drawn further into dynastic competition with South India, the life-cycle of political centres became shorter and shorter, but perhaps by that time extraordinary continuity had done its work. In short, Sri Lanka’s geographical position may have provided it with enough ‘protection’ for profound sensations of continuity to develop, and enough ‘exposure’ or alterity for self-conscious identities to form around those continuities.^[97]

This may seem like a very abstract analysis, but in one sense it does find an echo in the perceptions of historical Sinhala speakers themselves. As we saw, from the viewpoint of the courts in the Sinhala-dominated south, the island has long been a distinctive self-contained arena of political aspiration. That the ‘reification’ of Lanka has a considerable history before colonial powers ever started to imagine the land for themselves is worth emphasizing given a common tendency to emphasize British ideological manufacturing in particular. Nevertheless a certain puzzle remains. Was the rest of the Sanskrit cosmopolis really so divergent from the European trajectory as Pollock claims? If so, why should the Pāli land-scape take on contours that look a little more ‘European’ in appearance–at least when viewed in a certain light? Was the Lankan crucible really so important?

Footnotes and references:

[back to top]

[1]:

Sheldon Pollock, The Language of the Gods in the World of Men: Sanskrit, Culture, and Power in Premodern India (Berkeley and Los Angeles: University of California Press, 2009), 12.

[2]:

Pollock, Language of the Gods, 26, on the ‘cosmopolitan vernacular–that register of the emergent vernacular that aims to localize the full spectrum of literary qualities of the supposed cosmopolitan code’.

[3]:

Pollock, Language of the Gods, 28.

[4]:

Also see Charles Hallisey, ‘Works and Persons in Sinhala Literary Culture,’ in Literary Cultures in History: Reconstructions from South Asia, ed. Sheldon Pollock (Berkeley and Los Angeles: University of California Press, 2003), 721.

[5]:

He goes on: ‘though important kāvya in the language was produced there and power did sometimes express itself in Sanskrit as late as the mid-ninth century. Yet Sanskrit was a shaping force behind the scenes in Sri Lanka and elsewhere in the Pāli world.’ Pollock, Language of the Gods, 386. See Gornall and Henry in this volume on the development of Pāli as literary language under Sanskritic influence (Chapter 4).

[6]:

Pollock, Language of the Gods, 399.

[7]:

Pollock, Language of the Gods, 387.

[8]:

Pollock, Language of the Gods, 413. On Lankan understandings of the term cakravarti in this period see John Clifford Holt, The Buddhist Viṣṇu: Religious Transformation, Politics and Culture (New York: Columbia University Press, 2004), 41, and S. Pathmanathan, ‘South India and Sri Lanka, AD 1450–1650: Political, Commercial and Cultural Relations,’ Journal of Tamil Studies 21 (1982), 36–57.

[9]:

Pollock, Language of the Gods, 390.

[10]:

The most significant example being the expeditions sent to Southeast Asia by Parākramabāhu I (1153–86). Jonathan Walters, ‘Buddhist History: The Sri Lankan Pāli Vaṃsas and their Community,’ in Querying the Medieval: Texts and the History of Practices in South Asia, ed. Ronald Inden, Jonathan Walters and Daud Ali (New York, 2000), 141, argues that the Okkaka dynasty’s claims to dominance over South Indian dynasties was a serious expression of the political imagination. See also Pollock, Language of the Gods, 420, and Berkwitz’s discussion in this book of the geography of the fifteenth- century Sandēśa poems (Chapter 5).

[11]:

Holt, Buddhist Viṣṇu, 42–3.

[12]:

R. A. L. H. Gunawardana, ‘The People of the Lion: The Sinhala Identity and Ideology in History and Historiography,’ in Sri Lanka: History and the Roots of Conflict, ed. Jonathan Spencer (London and New York: Routledge, 1990), 61. Also see Hallisey, ‘Works and Persons in Sinhala Literary Culture’, 733.

[13]:

Strathern, Kingship and Conversion, 127–28. 92

[14]:

See Lorna Dewaraja, ‘The Indigenisation of the Muslims of Sri Lanka,’ in Sesquicentennial Commemorative Volume of the Royal Asiatic Society of Sri Lanka, 1845–1 995, ed. G. P. S. H. de Silva and C. G. Uragoda (Colombo: Royal Asiatic Society of Sri Lanka, 1995), 427–3 9.

[15]:

Also see Elizabeth Nissan and R. L. Stirrat, ‘The Generation of Communal Identities,’ in Sri Lanka: History and the Roots of Conflict, 19–44, which is influenced by Gellnerian modernization theory.

[16]:

R. R. Davies, ‘The People of Britain and Ireland 1100–1400. I. Identities,’ Transactions of the Royal Historical Society, Sixth Series 4 (1994), 11. Indeed, premodern polities in general contained an ‘imperial’ propensity to rule over others. The distinction between imperial and vernacular polities cannot therefore be pushed too far.

[17]:

See Gunawardana, ‘People of the Lion’, 69, and Michael Roberts, ‘Ethnicity after Said: Post-Orientalist Failures in Comprehending the Kandyan Period of Lankan History,’ Ethnic Studies Report 19 (2001), 69–98, for two contrasting views.

[18]:

Pollock, Language of the Gods, 474, see also 505–10, and Sheldon Pollock, ‘Cosmopolitan and Vernacular in History’, in Cosmopolitanism, ed. Sheldon Pollock, Homi K. Bhabha, Carol Appadurai Breckenridge and Dipesh Chakrabarty (Durham, NC and London: Duke University Press, 2002), 36–7.

[19]:

Pollock, Language of the Gods, 476, 510.

[20]:

Pollock, Language of the Gods, 475. Although Pollock refers to ‘Indian’ texts here, in the same paragraph he refers to ‘South Asian’.

[21]:

Pollock, Language of the Gods, 276. Shared descent is emphasized in the list of features of ethnicity in Anthony D. Smith, National Identity (London: Penguin, 1991), 21–2.

[22]:

Davies, ‘People of Britain and Ireland’, 4.

[23]:

Davies, ‘People of Britain and Ireland’, 6.

[24]:

John Watts, The Making of Polities: Europe, 1300–1500 (Cambridge: Cambridge University Press, 2009), 141–3.

[25]:

As Pollock, Language of the Gods, 511, says, the great Indian regional narratives are of kings and places–to which we must add in the Lankan case, monastic orders.

[26]:

See Berkwitz, Chapter 5 in this volume and Alan Strathern, ‘Towards the Source-Criticism of Sītāvakan Heroic Literature. Part 2: The Sītāvaka Haṭana: Notes on a Grounded Text,’ Sri Lanka Journal of the Humanities 34 (2008), 45–72.

[27]:

Pollock, Language of the Gods, 475–6, 510. See Gunawardana, ‘People of the Lion’, 63, for evidence of the rigidity of caste (jāti) distinctions in the eleventh century, for example. See also John D. Rogers, ‘Caste as a Social Category and Identity in Colonial Lanka,’ The Indian and Economic Social History Review 41 (2004), 51–77.

[28]:

At one point Pollock, Language of the Gods, 511, indicates that this definition is only according to ‘current scholarly opinion’, and he shows little inclination elsewhere to tag along where current scholarly opinion leads. Yet alternative definitions of ethnicity or ‘peoplehood’ are not explored.

[29]:

Davies, ‘People of Britain and Ireland’, 4.

[30]:

Obeyesekere, ‘On Buddhist Identity’, 230.

[31]:

Davies, ‘People of Britain and Ireland’, 9, on Ireland in particular, which had a much weaker all-island regnal tradition than Lanka. Otherwise it is a good comparator for Sri Lanka given that it is roughly the same size and also endowed with a form of hierarchical kingship and awkward relations with overbearing neighbours. Indeed, according to Patrick Wadden, ‘Ethnic or National? Identity in early Medieval Ireland,’ paper at the ‘Identity, Ethnicity, and Nationhood before Modernity’ conference (see footnote 1), Irish common identity persisted despite a far weaker record of actual political attainment of unity than we find in Lanka.

[32]:

Davies, ‘People of Britain and Ireland’.

[33]:

Including features that Pollock identifies beyond fictive kinship, for example, ‘a discourse that links language, identity, and polity’. Michael Roberts has published extensively on Sinhala ethnicity. His latest account is ‘Sinhalaness and its Reproduction, 1232–1 818,’ in The Sri Lankan Republic at Forty: Reflections on Constitutional History, Theory and Practice, Vol. 1, ed. Asanga Welikala (Colombo: Centre for Policy Alternatives, 2012), 253–87.

[34]:

Mahāvaṃsa, The Great Chronicle of Sri Lanka, trans. Ananda W. Guruge (Colombo: Associated Newspapers of Ceylon, 1989), chapters 6–7; The Dīpavaṃsa, trans. Hermann Oldenburg (1879; reprint New Delhi: Asian Educational Services, 2004), chapter 9.

[35]:

See Alan Strathern, ‘The Vijaya Origin Myth of Sri Lanka and the Strangeness of Kingship,’ Past and Present 203 (2009), 3–28. 293

[36]:

Strathern ‘The Vijaya Origin Myth’. Since writing the latter I have come across Peter Schalk, ‘Referents and Meanings of Siṃhala/ Sīhaḷa/ Ciṅkaḷam,’ in Kontinuitäten und Brüche in der Religionsgeschichte, ed. Michael Stausberg (Berlin and Boston: De Gruyter, 2012), but it does not bear centrally on the main arguments here.

[37]:

Gunawardana, ‘People of the Lion’, and later, Historiography in a Time of Ethnic Conflict: Construction of the Past in Contemporary Sri Lanka (Colombo: Social Scientists’ Association, 1995); K. N. O. Dharmadasa, Language, Religion and Ethnic Awareness: The Growth of Sinhalese Nationalism in Sri Lanka (Michigan: University of Michigan Press, 1992).

[38]:

Gunawardana, Historiography, 35–51. The phrase ‘public transcript’ is taken from James C. Scott, Domination and the Arts of Resistance: Hidden Transcripts (New Haven: Yale University Press, 1990).

[39]:

Gunawardana, Historiography, 41; compare Dharmadasa, Language, Religion, 20.

[40]:

Gunawardana argues that the text should be read as referring to (some) residents of the land; Dharmadasa to (the) residents of the land. Yet if one grants the former, this may simply refer to the fact that the Sinhalas shared the island with Tamil-speakers and väddās and others. The tenth-century Vaṃsatthappakāsinī says that the 700 followers and all their descendants ‘up to the present day’ are called Sinhala. Gunawardana argues that this implies the exclusion of the service castes whose arrival is also mentioned. But see Alan Strathern, ‘Vijaya and Romulus: Interpreting the Origin Myths of Sri Lanka and Rome,’ Journal of the Royal Asiatic Society of Great Britain and Ireland 24 (2013), 12–13, for a different interpretation of the significance of the arrival of the service castes in the Mahāvaṃsa. See Hallisey, ‘Works and Persons in Sinhala Literary Culture’, 730, for a tenth- century Sanskrit reference to the ‘Island of the Sinhalas’.

[41]:

Pollock, Language of the Gods, 474.

[42]:

In the Dharmapradīpikā, the founding exploit of the dynasty gives the latter its name, this extends to the land itself, the land then gives its name to the people who reside in it. Gunawardana, ‘People of the Lion’, 64; Gunawardana, Historiography, 53, argues that this represents the shift to the more inclusive sense of Sinhala that he does not see in the Dhampiyā Aṭuvā Gäṭapadaya and even says (p. 58) that ‘the process leading to this change would have very well started at an earlier time’.

[43]:

Pollock, Language of the Gods, 387, concedes in an equivocal fashion: ‘Some scholars also suggest that it was at this time that the term “Sinhala” first acquired more noticeable if still somewhat vague connotations of a wider political community, no longer referring, as it had earlier, merely to ruling lineages.’

[44]:

As a testament to Lanka’s participation in a cosmopolitan Buddhist order, Xuanzang seems to have picked up his information from Lankan monks who had come to stay in Kanchipura in Southern India, see Hui Li, The Life of Hiuen-Tsiang (1911; reprint Delhi: Munshiram Manoharlal, 1973), 139. Yet he may have digested this by means of a Chinese discourse of essentializing barbarian peoples. Xunazang’s account differs in certain details from the Mahāvaṃsa account; indeed it belongs to what I identified as the ‘Sinhala’ strain in Strathern, ‘Vijaya and Romulus’, see the table on p. 9. It is odd that Peter Schalk, ‘Referents and Meanings’, 555, discusses Xunzang but not these details.

[45]:

I am very grateful to Peter Ditmanson, of the Faculty of Oriental Studies at Oxford University, for translating this section of the text on my behalf. In ‘Vijaya Origin Myth’, 28, I referred briefly to this text and there relied on the translation in Si-Yu-K i: Buddhist Records of the Western World, Translated from the Chinese of Hiuen Tsiang (AD 629), trans. Samuel Beal (Delhi: Motilal Banarsidass: 1994), 240, which is not literal enough for our purposes here (the term ‘race’, for example, has no equivalent in the text).

[46]:

My thanks to Jeff rey Sundberg for bringing to my attention its use in texts and inscriptions from Lanka, Java and Chinese travellers, particularly around the eighth century. See, for example, Jeff rey Sundberg in collaboration with Rolf Giebel, ‘The Life of the Tang Court Monk Vajrabodhi as Chronicled by Lü Xiang: South Indian and Śrī Laṅkān Antecedents to the Arrival of the Buddhist Vajrayāna in Eighth-Century Java and China,’ Pacific World, Third Series 13 (2011), 204, n. 135, and 207, n. 137; and Jeffrey Sundberg, ‘The Abhayagirivihāra’s Pāṃśukūlika Monks in Second Lambakaṇṇa Śrī Laṅkā and Śailendra Java: The Flowering and Fall of a Cardinal Center of Influence in Early Esoteric Buddhism,’ Pacific World, Third Series 16 (2014), 119, n. 20; and H. Sarkar, ‘The Kings of Sri Sailam and the foundation of the Sailendra Dynasty of Indonesia,’ in Bijdragen tot de Taal-, Land-en Volkenkunde 141 (1985), 323–38.

[47]:

See Hallisey, ‘Works and Persons in Sinhala Literary Culture’, 698, for an inscription by the eleventh-century king Mahinda IV.

[48]:

Apart from the Pūjāvaliya, see Stephen C. Berkwitz, The History of the Buddha’s Relic Shrine: A Translation of the Sinhala Thūpavaṃsa (New York: Oxford University Press, 2007), 187–91. 294

[49]:

On war as the shaper par excellence of group self-identity, see Davies, ‘People of Britain and Ireland’, 10.

[50]:

See Gananath Obeyesekere, ‘Buddhism, Nationhood and Cultural Identity: A Question of Fundamentals,’ in Fundamentalisms Comprehended, ed. Martin E. Marty and R. Scott Appleby (Chicago: University of Chicago Press, 1995), 239–44.

[51]:

See also Marshall Sahlins, ‘The Stranger-K ing: or, Elementary Forms of the Political Life,’ in Stranger-Kings in Indonesia and Beyond, ed. Ian Caldwell and David Henley, special issue of Indonesia and the Malay World 36 (2008), 177–99.

[52]:

They represent the Sinhalas as an ethnic group possessing particular natural characteristics; a certain religious system, literary tradition, script, set of customs and also ancestry. Naturally, we must always be alert to the imposition of European understandings on these stories, but see the discussion in Strathern, Kingship and Conversion, 236–40.

[53]:

Fernão de Queyroz, The Temporal and Spiritual Conquest of Ceylon, trans. and with introduction by S. G. Perera, 3 vols. (1931; reprint New Delhi: Asian Educational Services, 1992), Vol. 1, 7. See also, for example Diogo do Couto in The History of Ceylon from the Earliest Times to 1600 AD as Related by João de Barros and Diogo do Couto, ed. and trans. D. Ferguson (New Delhi: Asian Educational Services, 1993), 65, 101.

[54]:

See Alan Strathern, ‘Treachery and Ethnicity in Portuguese Representations of Sri Lanka,’ in Engaging Colonial Knowledge: Reading European Archives in World History, ed. Ricardo Roque and Kim Wagner (Basingstoke: Palgrave Macmillan, 2012), 217–3 4. Note that Pollock (e.g. Language of the Gods, 511, n. 21) is more prepared to see ethnicity emerging in some parts of India during the early modern period.

[55]:

Hallisey, ‘Works and Persons in Sinhala Literary Culture’, 691.

[56]:

Walters, ‘Buddhist History’, 135–6.

[57]:

Pollock, Language of the Gods, 278–80. It is not always clear whether and when ‘Sanskrit Cosmopolis’ and ‘India’ are intended as synonyms–or if the latter was carefully chosen to avoid Lanka.

[58]:

Pollock, Language of the Gods, 279.

[59]:

Mahāvaṃsa, 1.84, on the three visits to Lanka of Buddha. Guruge’s translation is ‘the island, honoured by the pious, came to be resplendent as the righteous isle’. This has often been glossed as rendering Sri Lanka dhammadīpa (island of the dhamma). Steven Collins, Nirvana and Other Buddhist Felicities. Utopias of the Pāli Imaginaire (Cambridge: Cambridge University Press, 1998), 598–9 (whose translation is ‘the island made radiant with the light of the Dhamma, became highly respected by all good people’), and Walters, ‘Buddhist History’, 147, argue that this gloss is recent and carries inappropriate connotations of exclusivity. But an explicit claim to exclusivity need not be an element of European Providentialism either. What is crucial and common to both is the idea of a territory/ society/ polity as an exemplary and leading force in the establishment of a religious dispensation, which will reap both earthly and salvific rewards as a result. Moreover, see below on an intriguing verse from the Sītāvaka Haṭana of 1585.

[60]:

See for example Obeyesekere, ‘On Buddhist Identity’, 229. For an analysis of the functioning of national Providentialism in medieval England, see Andrea Ruddick, ‘National Sentiment and Religious Vocabulary in Fourteenth-Century England,’ The Journal of Ecclesiastical History 60 (2009), 1–18.

[61]:

Mahāvaṃsa, 23.7–17. Berkwitz, Thūpavaṃsa, 179; and see Gornall and Henry in this volume (Chapter 4).

[62]:

Mahāvaṃsa, 23.8–15.

[63]:

Berkwitz, Thūpavaṃsa, 166. Note that at one point the Sangha is described as the family deity [kuladēvatā] of the royal family (p. 164).

[64]:

The fourteenth-century Saddharmalankaraya has Duṭugāmaṇu announce ‘it is time to rid the country of the Tamil scum and to purify the Dispensation of the Buddha’ as translated in An Anthology of Sinhalese Literature up to 1815, ed. C. H. B. Reynolds (London: George Allen and Unwin, 1970), 245, and also see the account in the seventeenth-century Rājāvaliya, ed. and trans. A. V. Suraweera (Ratmalana: Vishva Lehka, 2000), 35, 41.

[65]:

Sheldon Pollock, ‘Rāmāyaṇa and Political Imagination in India,’ The Journal of Asian Studies 52 (1993), 261–97.

[66]:

See Victor Lieberman, Strange Parallels: Southeast Asia in Global Context c.800–1830, Vol. 2: Mainland Mirrors: Europe, Japan, China, South Asia and the Islands (Cambridge: Cambridge University Press, 2009), 43, on how universal religions such as Theravāda Buddhism may fuel ethnic particularism and disdain.

[67]:

Mahāvaṃsa, 36.41, 36.110–11; Amaradasa Liyanagamage, State, Society and Religion in Premodern Sri Lanka (Colombo: Social Scientists,’ Association, 2008), 121–3 3; John Cliff ord Holt and Sree 295 Padma, ‘Buddhism in Andhra and its Influence on Buddhism in Sri Lanka,’ in Buddhism in the Krishna River Valley of Andhra, ed. Sree Padma and A. W. Barber (Albany, NY: SUNY Press, 2008), 116.

[68]:

Hallisey, ‘Works and Persons in Sinhala Literary Culture’, 702; R. A. L. H. Gunawardana, Robe and Plough: Monasticism and Economic Interest in Early Medieval Sri Lanka (Tucson: University of Arizona Press, 1979), 172.

[69]:

Charles Hallisey, ‘Devotion in the Buddhist Literature of Medieval Sri Lanka’, PhD dissertation, University of Chicago, 1988.

[70]:

Walters ‘Buddhist History’, 133. ‘Theravāda’ has been placed in quotation marks here because before the nineteenth century it carried rather different meanings, as discussed in Peter Skilling, ‘Theravāda in History,’ Pacific World, Third Series 11 (2009), 61–93. Nevertheless, whatever its changing emic contents, it remains a reasonable etic term to describe the particular form of Buddhism dominant in Lanka and Southeast Asia.

[71]:

Hallisey, ‘Works and Persons in Sinhala Literary Culture’, 701.

[72]:

See also Gunawardana, Robe and Plough, 177.

[73]:

Obeyesekere, ‘On Buddhist Identity’, 234, has given us the following translation of an excerpt: ‘Just as the demons could not find permanence here, neither can this land become a place of residence for non-believers (mityādṛṣṭi gatavungē vāsaya). If any non-believer becomes a king of Sri Lanka by force at any time, that dynasty will not last owing to the special influence of the Buddha. Because this Lanka is rightfully those of kings who have right views [Buddhists], their rightful dynastic ten- ure (kula pravēṇiya) will absolutely prevail. For these various reasons the kings of Sri Lanka are drawn by a natural love of mind to the Buddha and will establish the sāsana without delay or neglect and protect the wheel of the law and the wheel of doctrine and reign so that the rightful dynastic tenure will be preserved.’

[74]:

Cūḷavaṃsa: Being the More Recent Part of the Mahāvaṃsa, ed. and trans. Wilhelm Geiger, 2 vols. (1925–9; reprint London: Pali Text Society, 1980), 80.54–7 0; Liyanagamage, State, Society and Religion, 24.

[75]:

The phrase is from the Upāsakajanālaṅkāra (The Adornment of the Buddhist Laity), which Liyanagamage, State, Society and Religion, 19, dates to the thirteenth century.

[76]:

Liyanagamage, State, Society and Religion, 255–65, places this in the context of some other lapses from the usual Indic pattern of religious tolerance in the twelfth century (Kashmir and South India).

[77]:

Pollock, Language of the Gods, 430–2, 479, refers to Karnataka (twelfth and thirteenth centuries) and Gujarat (fifteenth century).

[78]:

One might consider here the Dahampiyā Aṭuvā Gäṭapadaya, a Sinhala glossary of Pāli terms with an exegetical function, from the ninth century. However this may relate to scholarly activity with little popularizing purpose, Stephen Berkwitz, personal communication, 7 June 2013.

[79]:

Alan Strathern, ‘Sri Lanka in the Long Early Modern Period: Its Place in a Comparative Theory of Second Millennium Eurasian History,’ Modern Asian Studies 43 (2009), 809–64, which cites Walters (personal communication, 21 June 2004) on these early vernacular centuries as pivotal for ‘the deep integration of Buddhism into popular consciousness and practice’, the establishment of ‘domesticated Buddhism’. See also Gornall and Henry in this volume, on Sinhala texts with a lay audience in mind developing alongside the adoption of Sanskritic aesthetic norms from the twelfth century (Chapter 4).

[80]:

Hallisey, ‘Works and Persons in Sinhala Literary Culture’, 707, 732–3 7.

[81]:

Berkwitz, Thūpavaṃsa, 7–12, 17–20. Berkwitz (pp. 20–2), suggests that it introduces dramatic moments into its Pāli source in order to captivate its audience.

[82]:

Stephen C. Berkwitz, ‘The Expansion of Buddhism in South and Southeast Asia,’ in Dynamics in the History of Religions between Asia and Europe: Encounters, Notions, and Comparative Perspectives, ed. Volkhard Krech (Leiden: Brill, 2011), 228.

[83]:

Pollock, Language of the Gods, 490–4; Pollock, ‘Rāmāyaṇa’.

[84]:

Pollock is prevented from pursuing further such suggestions as ‘war may have made vernacular poli- ties’ by an aversion to functionalism.

[85]:

Strathern, Kingship and Conversion, 168–73, 196–7.

[86]:

Of course, legitimation theory may be an inappropriate–and certainly an insufficient–prism through which to perceive a whole range of cultural artefacts. The logical misstep is to claim that it is irrelevant to all cultural production within a certain society or civilization. In short, if it would be perverse to read Shakespeare’s Richard II (cited below) as merely an exercise in royal legitimation, it would hardly do to conceive of Ben Johnson and Inigo Jones’ masques as artistic exercises unrelated to the projection of power. As the case of the masques well illustrates, however, such exercises rarely simply flatter power but also seek to shape, constrain and critique it. 296

[87]:

Pollock, Language of the Gods, 18, 257–58, 511–31 on legitimation theory. A vast accumulation of insight from the social sciences (not just the old targets of Marx and Weber, but Bourdieu and Foucault to James Scott) is thereby set aside–but also misconceived. Pollock assumes that legitimation entails the deployment of conscious strategies and even deceptions by ruling elites. Although some innovations of sacred kingship imagery surely involved an element of calculation, such as Niśśankamalla’s proclamation above, in general it is quite wrong to imagine that the production and reproduction of ideology depends on the insincerity of individual utterances or the adoption of some disenchanted locus of perspective above the society in question. In the end, Pollock fails to find a route between the Scylla of ‘total state ideology’ and the Charybdis of the ‘theatre state’ (p. 257) as he terms it. At one point he does finally make an analytical connection between culture and power by suggesting that literature worked towards the ‘ennoblement of political life’ (p. 524)–which is surely to invite a version of legitimation theory through the back door. It is worth noting, however, that the religious and literary embellishment of kingship is rarely only a production for subject consumption but is also aimed at establishing a certain status with regard to other rulers: as we have emphasized, much court culture is a function of peer rivarly.

[88]:

It makes sense to connect the ‘constant struggle for autonomy’ in the literary material, as Hallisey, ‘Works and Persons in Sinhala Literary Culture’, 699, puts it, with the struggle for autonomy in political terms.

[89]:

I am not in a position to judge the extent to which Pollock is right about the rest of the Indic world.

[90]:

Collins, Nirvana, 72.

[91]:

See Tilman Frasch, Chapter 3 in this volume.

[92]:

In her recent and stimulating ‘Wilde Lectures in Natural and Comparative Religion: Defining Buddhist Formations Across the Indian Ocean,’ Oxford University, May–June, 2016, Anne Blackburn made the case for setting aside the term ‘Theravāda’ (as having a quite diff erent emic resonance in this period) in favour of the term ‘Pāli-o riented Buddhism’, and also warned against assuming that there was any one-directional export of a coherent ‘package’ from Lanka to Southeast Asia. I have not sought to draw on her arguments properly here but note that they may help us in the future to reframe how we consider these and related issues explored in the introduction.

[93]:

Lieberman, Strange Parallels.

[94]:

Previous generations of historians have often stressed its island status, and not always with crudely nationalistic aims. See R. A. L. H. Gunawardana, Periodization in Sri Lankan History: Some Reflections with Special Emphasis on the Development of the State (Colombo: Social Scientists’ Association, 2008), 20.

[95]:

This point was also made in Strathern, ‘Sri Lanka in the Long Early Modern Period’. Note, however, that the north of the island was ruled as a province of the Cōḻa empire for seventy-s even years after the fall of Anurādhapura, 993–1070 ce.

[96]:

Pollock, Language of the Gods, 411.

[97]:

See Jonathan S. Walters, ‘Mahāsena at the Mahāvihāra: On the Interpretation and Politics of History in Pre-Colonial Sri Lanka,’ in Invoking the Past: The Uses of History in South Asia, ed. Daud Ali (New Delhi: Oxford University Press, 1999), 338–41, for a strong emphasis on the consciousness of over- bearing neighbours (sometimes lieges, sometimes conquerors) through the Anurādhapura period and beyond.