Michael Roberts’ writings have sometimes given the impression of a man who will write at the drop of a hat and at great speed: the subjects have been many and various; the approach as openly adversarial as many of the relationships he takes as his subject; the arguments occasionally advanced by death-defying conceptual leaps or obscure symbolic readings; the prose style quirky or impatient with the more conventional norms of academic prose. The latter is evident even in the present work, in fact the culmination of decades of reflection, where he refers openly to his own intellectual progress, to arguments with colleagues, even to his own ethnic category – Tuppahiyek, or ‘mongrel’ – and sees no cause for shame in routinely citing ‘personal communication’ or telephone conversations in is footnotes. Such considerations might induce the superficial reader to underestimate the importance of the arguments presented in this new monograph. In fact it deserves to be widely read by all those interested in the vigorous debates about ethnic sentiment, nationalism and the murky passage from one to the other.

Its chief strength is the author’s appreciation of what kind of thing group identity is, a wisdom that comes from a lifetime of being alert to the implications of affiliation and status and the perversions and paradoxes entailed in living them out. Its chief contribution is the force with which it sets out the case that the Sinhalese were conscious of their ‘Sinhalaness’ well before the interventions of modernity or British colonialism, that this consciousness was manifest in a xenophobic discourse which looked to distant history in order to conflate external enemies past and present, and that this succeeded, in combination with a sense of place and political loyalty, in eliciting a form of patriotism or proto-nationalism. The chief corpus of primary evidence on which this case rests is a number of hatan kavi, or war poems, from the late-sixteenth to eighteenth centuries, most of which have been published and used for many decades but which have escaped theoretically-informed critical attention until now. They are certainly robust enough to support Robert’s central claims. In the contextualization of them, however, the book is more theoretically persuasive than empirically irrefutable.

That is to say, the evidence is sometimes not substantial enough to support suggestions made while embedding these poems and the other key sources in their cultural setting. Roberts is not remotely coy about this, regularly admitting to speculation, even ‘wild speculation’. Indeed it is an admirable feature of the book that he is so frank in laying bare the mechanics of scholarship: he is pretty scrupulous in noting divergences of opinion, weaknesses in his argument, places where he disagrees with his sources or translators. At some point, the reader ought to realize that if Roberts is adversarial, it is only in the most sportsmanlike of ways. One feels invited to disagree, and the rest of this review will take up that invitation while offering support to the book’s overall contribution and conclusions.

What kind of speculations do we get? Roberts does not say so explicitly, but his theoretical inclinations seems to be towards symbolic anthropology of a Geertzian stamp, as his reference to an emphasis on ‘symbolic webs of meaning’ (p.16) would indicate. A strong form of holism is at work here, so that patterns or symbols from one cultural domain are expected to be replicated in another. This works very well when meditating on the intense transferences between divine and secular hierarchies, or when decoding the details of a ceremony such as the royal Asala Perahara, his account of which is very reminiscent of Geertz’s work on the Balinese ceremonies of kingship. Inevitably, the analysis is sometimes troubled by the dilemma faced by all who have tried to transfer this method to history: how to ‘read the signs over the native’s shoulder’, when the native and his people are now long gone. In the desire to recreate even a fraction of worlds gone by, textual fragments and clues can be hoovered up from across great expanses of time.

In short, one criticism that Roberts can expect to face from those so inclined is that of ahistoricism. Again, he is utterly open about this, acknowledging several times that he is treating 1590s to 1815/8 as a single unit, without attempting to introduce a sense of change over time. In his defence one could say that there is a certain coherence to this epoch (which I would push back to the 1550s: see below), characterized as it was by a militiarization of society, by rebellion against and incorporation of European occupiers of the lowlands – the situation of many other societies of ‘early modern’Asia. Moreover diachrony is difficult to come by on such questions of identity because of the fragmentary, partial and spasmodic nature of the relevant evidence. Nevertheless, its absence is a weakness, particularly when he extends his reflections back to include what he refers to as the whole of the ‘middle period’ (1232-1818 A.D.).

That Roberts – or any writer on Sri Lankan history – can expect stiff criticism is beyond doubt. He provides excellent background for those interested in the hot-blooded debates in Lankan studies, inflamed by gusts from the subcontinent and the global winds of intellectual fashion that have blown across the island’s past. The latter are diverse, somewhat contradictory: the strange popularity of instrumentalist approaches to ethnicity in Anthropology (strange because of that discipline’s typical disgust of reductionism); the post-modern taste for the amorphous and the shifting, in which all identities – even concepts – can become as wobbly and unreliable as jelly-fish; the much-discussed assertions of the modernist or modernization theorists (most famously Ernest Gellner and Benedict Anderson).  What comes out of Roberts’ work are the problems afflicting three common assumptions of the modernists’ arguments: (1) the notion that pre-modern modes of communication and mobility were so weak as to be incapable of producing forms of solidarity or homogeneity at any but the most local of levels, (2) the notion that a certain kind of equality and homogeneity are necessary conditions of patriotic identity and solidarity, and (3) the notion that extensive traffic across ethnic boundaries necessarily indicates the political irrelevance of those boundaries.

Communication:

Chapter Two in particular sets out to show how the Sinhala and Pali texts are often best seen not as isolated islands of elite and partisan argument but as the visible tips of great icebergs of oral and popular communication plunging downwards largely beneath our view. In doing so we are provided with a fascinating excursion into the pre-modern media that might allow common ideas and values to be spread and sustained throughout a kingdom. This represents a timely injection into Lankan studies of the insights from a great mass of recent anthropological work on cultural transmission, and also brings to mind, in its evocations of how word can be spread to the illiterate through lecture, gossip and image, Bob Scribner’s work on Reformation Europe. As Roberts discusses how inflexible oral compositions can be when stiffened with the starch of alliteration and repetition, how written and oral utterance interweave and mimic each other, how much can be transmitted and embodied in dance and song or represented in temple murals and architecture, it is as if we are being reminded of a priori facts about the nature of communication itself, which we only need reminding of because of modernization theorists’ focus on the primacy of the printed word.

However, the evidence on the prevalence and reception of these media in early modern Sri Lanka is not always firm enough to withstand the anticipated stampings of skeptics. We need more information, for example, on whether poetry competitions were more than a metropolitan diversion, or on whether all classes and regions took part in pilgrimages. Then again, it is precisely that kind of evidence that is least likely to survive. We may have to behave like classicists, recreating a society from a pottery-shard. An insight into the way in which songs could shape something that looks a bit like ‘public opinion’ is afforded by the Sinhala chronicle Rajavaliya. We are told  of a quarrel between an Indian military officer who was a favourite of Rajasinha I  (r. 1581-93) of Sitavaka,  and a rival courtier, which ended in the latter composing four-lined verses about the Indian and forcing him to shift residence by having ‘got these verses to be recited in Sitavaka’, (p. 24). Even when he was summoned back to the city by the queen, ‘lads began to sing these verses, whenever he was seen.’ Strangely, given his agenda, Roberts does not emphasize that the latest translation of the Rajavaliya suggests that this campaign was waged against him ‘for the reason that he was having company with fakirs’.[1]

Even better is a piece of evidence from the abnormally perspicacious and sensitive Robert Knox, long-term prisoner of the King of Kandy, on the flourishing institution of ambalamas (stations for wayfarers) and the conversations they sustained in the seventeenth century. Knox’s comments are so headily evocative of how men could participate in a ‘public sphere’ revolving around a powerful monarch that Roberts quotes it twice and I cannot resist following suit:

‘At their leisure when their affairs permit, they commonly meet at places built for strangers and way-faring men to lodge in called Amblomb, where they sit chewing Betel… discoursing concerning the Affairs at Court, between the King and great Men; and what Employment the People of the City are busied about. For as it is the chief of their business to serve the King, so the chief of their discourse is concerning such matters. Also they talk of their own affairs, about Catel and Husbandry, Laws and Government of their Countrey…’ (pp. 27-8)

Is this the Lankan equivalent of the roughly-contemporary British coffee-house? As it happens references to ambalamas can be found in fifteenth-century poetry, but we would really need a statistical and geographic survey of them in order to appreciate how significant they were in knitting together the regions and society as a whole. Another piece of verificatory research would be even more valuable. If Roberts is correct in suggesting that the Sinhala language was remarkably homogeneous (lacking any major dialect variations) as far back as the thirteenth century, then this really would represent a serious breach in Hobsbawm’s argument that mass linguistic uniformity depends on a mass system of education.

As a source for this suggestion Roberts gives us a conversation he has had with a Lankan scholar, a quite typical procedure for chapter two. Indeed it would be fair to say that his respect for oral transmission extends to the production of citations! In this chapter, and wherever close linguistic work is involved, such as in analyzing the content of the war-poems, Roberts is reliant on a host of other scholars, mainly Sri Lankan. Indeed he tells us in a characteristically frank footnote that a version of this chapter was rejected by a referee for the Royal Asiatic Society of Sri Lanka on the grounds that it was all so well known – which nicely reveals the distance between foreign commentators and the hidden depths of Lankan academia, given how fresh it all seemed to me. The virtue of this collaborative method is that we are presented with a wealth of locally-held knowledge which has often remained unpublished, or at least undirected to the concerns of international academe. There are three problems, however. First, Roberts is naturally unable to evaluate their interpretations or translations for himself, lacking proficiency in this form of Sinhala. Since this reviewer shares that failing he is in turn unable to assess how deleterious that is! Roberts often ameliorates this situation by giving us a variety of translations. Nevertheless, a debate is needed and anticipated on such issues as the changing connotations of the term Trismhala. Much of the terminological work in chapter six in particular is exploratory, preliminary or inconclusive. He suggests but does not prove that the British used the term ‘Sinhalese Nation’ in a way that carried some modern nation-like connotations; and he suggests a number of Sinhala terms expressing we-ness or oneness that could have given rise to the term ‘nation’ as it appears in European translations of Sinhala letters, without proving that any one of them was actually used. There is no question however but that all this opens up the debate considerably.

Secondly, there are some points where we are given the opinions of experts as sources when it would have been much better to pursue the enquiry right back to the evidential base. It is surely an important assertion that from the eleventh century stories were usually written in way that enabled them to be chanted, so we ought to be told what these stories are and where to find them, rather than be directed to a letter the author has received (p. 24). For readers interested in change over the ‘middle period’, such citations can be frustrating. When we are told that from its composition in the thirteenth century the rather pugnacious and vamsa-esque Pujavaliya was chanted or read aloud at ceremonies, it is no consolation to be told merely that this information comes from a specialist whose authority is the sixteenth century. If the evidence on the use of the Pujavaliya in this way is early and solid enough it would indicate that Sri Lanka was in step with Southeast Asia in witnessing a characteristically ‘early modern’ religious development, rooting Buddhist values and concepts – and one might add, the patriotic assertions of the vernacular chronicles – in local lay society.

Lastly, Roberts’ respect for oral tradition means that he has an unusually generous approach to late sources, arguing, for example, that the tales of an illiterate twentieth-century bard can take us into the mentalities of the Kandyan period. Perhaps. But who knows? The most critical issue on which this has a bearing is that of the Nayakkar dynasty of Kandy, famously contested by R. A. L. H. Gunawardana, H. L. Senevirtane, and K. N. O. Dharmadasa. Was the abortive conspiracy of 1760 against Kriti Sri Rajasinha inflamed by a chauvinist objection to his Tamil origins? In offering support to this conjecture, the key account Roberts points to is the Sasanavitarna Varnanava, which tells us that the king was a Saivite. This text was written roughly eighty years later after the event but may well incorporate oral traditions going back to that time. Roberts’ frustration at the way in which a source such as this is dismissed out of hand is understandable, but he can only weaken his case by describing historian’s preference for the contemporary source as merely a ‘conventional preference’ and ‘hidebound’ reasoning when it is such a basic principle of source-criticism.

Roberts is on firmer ground in establishing that such anti-Saivite sentiments were quite possible at this time. It has been asserted that the kind of Sinhala Buddhist consciousness required to sustain this rivalry did not arrive until the modern era. This assertion is indeed thoroughly undermined by Young and Senanayaka’s recent publication, under the title The Carpenter-Heretic, of fascinating stories about heresy. This MS was inscribed only two years after the 1760 conspiracy and it must now be germane to any arguments about eighteenth-century identity, given that its theology betrays a demotic genesis – we are told that it emerged out of the milieu of non-textualized lay Buddhism – and the way in which it easily assimilates Christianity to long-standing tropes defaming and demonizing Saivism.  (As it happens the parameters of possibility were already clear from the work of Charles Hallisey who has highlighted the role of anti-Saivite and anti-Brahmanical discourses back into the first millennium). Incidentally, it does need clarifying that the editors state that ‘there is not the slightest tinge of anti-Tamil bias’ in these vociferously anti-Saivite stories.[2]  Indeed, however entwined the identities of Sinhala-ness and Buddistness undoubtedly are and were, they do need to be kept apart at times for analytical purposes. The above considerations, for example, may allow us to envisage a ‘powerful Buddhist faction’ behind the conspiracy, but not necessarily one with ‘indigenist leanings’ (p. 49).

Community:

Chapters three to five are devoted to fleshing out the kind of polity that could have produced the collective sentiments expressed by the war poems – Roberts is not going to get caught out by the de rigeur Foucaultian critique of power and knowledge! Here too he draws heavily on the work of scholars such as P. E. Pieiris, Lorna Dewaraja, Seneviratne, James Duncan, and D. G. B. de Silva, but, again, he angles all this material for the first time towards the big questions of the day. He is particularly concerned here with the indigenous or emic viewpoint. It is of course important that we understand that from the perspective of the Kandyan court the whole of Sri Lanka came under its canopy, and that we do not reduce the playing out of this ideology to mere coercion: no doubt it could cultivate genuine loyalties among peripheral chieftains and princes. Early nineteenth-century correspondence with the British bears out how heavy with significance the rites of tribute-presentation were to the Kandyan kings, so that they can be seen as the appropriate ends of political action in themselves. In the ethnographic work by James Brow and Deborah Winslow, Roberts finds an eloquent testimony to the way in which the image of kingship could become deeply integrated into the consciousness of remote villages. They found, for example, references to royalty in very local rural discourses about place-name origins or land-claims. Here anachronism might even work in Roberts’ favour, for such twentieth-century material on kingship imagery can only represent a persistence of ‘Kandyan period’ themes at a time when the colonial and post-colonial bureaucratic state was attempting to institute quite other political visions. This stands as a fine example of his talent for drawing out the theoretical pertinence of existing scholarship and making imaginative use of it.

Yet his emphatic championing of ideology over pragmatics can come to seem a rather artificial intervention. It is surely equally valid for historians to pursue an etic perspective as regards what all this entailed in terms of economic and political control, and surely possible to do this without being blinkered by ‘twentieth century notions of administrative authority’ (p.74). Indeed the concern of much recent scholarship on Asia, including the work of that arch-symbolist Geertz on Bali, has been to estrange us from the past precisely by pointing up the contrast between symbolic and pragmatic power. The presentation here runs the risk of eliding the dynamism that Southeast Asianists such as Stanley Tambiah have been so careful to build in to their models. In the sixteenth century, for example, there is no question but that the kings at Kotte imagined themselves heir to a cakravarti tradition of lordship over the island, but that did not stop Kandy attempting to break away and claim such titles for itself – contrary to what Roberts asserts in footnote 1 of chapter three. Again, Roberts’ objection to the use of the term feudalism to describe such arrangements does the job of highlighting the relative absence of secular principles of reciprocity between liege and vassal, but – given that the term has had so much specificity knocked out of it of late by revisionist histories of Europe – it is difficult to find its loose application by K. M. de Silva offensive. The latter uses it to refer simply to high levels of political decentralization and the assumption that producers were subject to an obligation of service as a condition for holding land. It might even help capture the behaviour of the Kandyan Kings in recognizing the jurisdictional rights of their Dutch vassals, or explain why Sinhalese, Portuguese and Dutch could all find a rough-and-ready common diplomatic language in the rites of homage or tribute presentation (dakum).

Now the modernists’ argument has been that such pre-modern political agglomerations could not support a widespread feeling of common identity or solidarity. Roberts simply asks, why? If the modern nation-state happens to involve, in Benedict Anderson’s phrase, a ‘deep horizontal comradeship’ (p. 101), does this mean that a sense of national or proto-national identity can only arise in this way? Does patriotism depend on an assumption of equality? Roberts demurs, and shows oneness in hierarchy. The war poems certainly stand as a vivid picture of what that looks like. Through chapters six and seven we arrive at a vision of an imagined community gathered under a figure whom at once symbolizes it, transcends it and protects it. We are made to think about war and what it can do to collective identity, how the regions outside the core are dragged in as battle-sites, as recruiting grounds. In peacetime this rajakariya, or king’s work was a means by which pre-modern Lankans routinely felt the touch of the state, but it now became an obligation decisive of life or death. In such situations, coercion is less effective than inspiration, the courage that comes from reflecting on the glorious victories of the past or lifting one’s gaze up to superhuman kings and superhuman gods. And these poems provide unambiguous evidence that all this amounted to a sense of patriotism expressed through xenophobic antipathies and a conviction of sovereign right to territory.

In other words the evidence is most firm where it most needs to be. What kind of primacy this discourse enjoyed is still up for debate. We cannot really provide a proper account of the circulation and reception of these poems, but what clues there are point firmly towards popular consumption: the generally simple nature of Sinhala, the fact that they exist in different recensions, suggestive of diverse creative reworking and appropriation over time, the earthily sexual and martial themes. Aesthetically, these verses strike one as the vocal equivalent of the masculine Kandyan dancing that one can see in the Perahara procession today. Buddhist Sinhalaness is only occasionally apparent, an important confirmation that the usual culprits of elite ideological production, the bhikkhus, cannot be collared here. At the same time, the imaginings of bhikkhus past certainly were at work, in echoes of the vamsa traditions.

Crossing boundaries:

Many scholars of Sri Lanka have reminded their readers of the cosmopolitanism of the island’s past, of how many diverse peoples have washed up on its shores and found acceptance. It might seem a bit habitual by now, akin perhaps to the way that Britain’s hybrid heritage has become a staple feature of all levels of history-production. But in Sri Lanka such a viewpoint is far from anodyne and is obviously a response to a Sinhalese nationalism that speaks only in terms of racial essentialism and antagonism. Nevertheless, we can be left with the impression that before the nineteenth century ethnic boundaries were so trampled underfoot that they were barely recognizable to the people rushing across them. Roberts, again, asks a simple question: where was this stampede headed? The evidence assembled to date shows it traveling one-way, with Tamil and Malayalam-speakers adopting Sinhala and not vice versa, which might suggest something about the status of the respective identities. This arises out of the substantial work by both Tambiah and Gananath Obeyesekere on the processes of ‘Sinhalization’ and Buddhicization’. Tambiah has in fact seen such subordinating incorporation as a ‘standard South Asian mode of differentially incorporating into an existing society sectarian or alien minorities’.[3]

It is on the contrariness and contradictoriness of identity that Roberts is best, slicing through the endless evocations of ‘fluidity’ to look at how conceptualizations of in-groups and out-groups, and the terms that express them, shift according to context.  The analysis of such ‘group shifters’ or ‘disemic’ terms ought to be intuitively grasped by all those who do not expect the rabid patriotism of an England supporter to be placed under existential threat on those Saturdays when he joins the rest of the Leeds United crowd in roundly abusing the good folk of Manchester. Roberts makes an instructive comparative foray into the development of Britishness in the same period, drawing largely on Linda Colley, which reveals how intermittent warfare between Britain and France could lead to the construction of an identity that papered over strong regional differences, and which involved much disparagement of the French at the very moment at which French manners were proving so appealing to the British aristocracy.

Indeed he suggests that it is precisely where two groups share a great deal, where mutual influence is high and intermingling common, that boundaries can become contentious and significant. This is a very fruitful idea for thinking about Sri Lankan history, beyond even his reflections on the middle period. For example, it strikes one that the two great moments of the post-Anuradhapura polities, the unifying reigns of Parakramabahu I (r. 1153-86) and Parakramabahu VI (r. 1411-66), represent the same sort of paradox: in both we see a reassertion of Sinhalese power against South Indian might at the same time that South Indian cultural influence was reaching a new high water-mark. In other words there is nothing odd about a society that voraciously draws in Hindu religious forms while generating virulent anti-Saivite discourses, or which (in theory) traditionally has royal families of foreign ancestry while hosting some chauvinist resentment of that fact among the native aristocracy. Indeed, whatever the particularities of the Nayakkar case, Roberts is right to emphasize that the ethnic identity of monarchs is always in some sense category-busting. A consideration of early modern English and British history, for example, would find the acceptance of monarchs or consorts of foreign blood a typical feature of the dynastic state, but one that rendered them neither incapable of arousing patriotic rhetoric nor immune to suspicions of their foreignness. One could even say that to be royal is always both to sum up the nation and yet to be foreign or set apart from the people who make it up.

What to do with a text or a pattern of violence:

There are many ways of thinking about the current debates in Lankan studies, but one way is to concentrate on hermeneutics. In which case, one side is made up of a group of scholars with a background in post-war anthropology and a corresponding appreciation of cultural particularity and persistence as carried forth in myth and symbol. Many of them, including the likes of Tambiah and Obeyeskere, are products of the fecund Peradeniya campus of the 1960s and 70s. On the other side stands a younger generation of scholars, broadly influenced by post-modernist currents and new forms of textual criticism, and understandably motivated by the desire to deconstruct the nostrums of chauvinism. The most exciting example here is Jonathan Walters. Of course, Roberts, a product of Peradeniya himself, belongs to the former camp, and comparisons with his anthropologist-cum-historian peers are not difficult to come by. If some find the synchronic approach in this work not to their taste they are unlikely to see the appeal of the structuralist modeling employed by Tambiah in World Conqueror and World Renouncer (Cambridge, 1976) in which a distinctively Theravadin form of state – the galactic polity – was presented as oscillating between two poles over two millennia.  And when Roberts repeatedly suggests that if other scholars fail to grasp the slippery business of identity it is because they are prisoners of twentieth-century epistemology (materialism/secularism/rationalism etc.), one is reminded of the special access to native mentality assumed by Obeyesekere in his famous quarrel with Marshall Sahlins.

These two groups, representing one aspect of the ‘traditionalist’ versus ‘historicist’ opposition as I described it in Modern Asian Studies 38:1, tend to look for different meanings in the source material for any historical enquiry. The danger of the former approach is that texts are automatically rendered reflections of fundamental cultural tropes, persisting mythic structures or fervently-held beliefs. The danger of the latter approach is that texts are automatically rendered rather dead political instruments, alive only as stratagems or as conversations with other texts. In the current climate the vigorous reiteration of the former approach makes for a welcome corrective, but one not without its problems. In his reading of the correspondence with the British, Roberts is right to draw our attention to the claims of divine protection made by the Kandyan kings, previously passed over as meaningless rhetoric but plausibly corresponding to actual contemporary feelings sustained by the mountain kingdom’s remarkable record in preserving its independence from European assault. However it is going too far to say the whole Sinhala elite were simply unperturbed by the British threat as a consequence. We do at least need an awareness of how the royal epistle had its own rhetorical world in which bombast and boast had a special diplomatic function. Equally the argument that the mightiness of the king’s image, and his divine, or god-like status, amounted to much more than court theatrics is well taken, but needs to be placed alongside the fact that this hardly diminished the Sinhalese’s great capacity for revolt and regicide when material interests were at stake. These ‘gods’ could be readily dispatched, if with some ritual disquiet. As for the significance of the war-poems themselves, it is an intriguing suggestion – backed up by just one allusion in the Rajasinha Hatana – that they had a constitutive or illocutionary function. That is to say, that they were considered not just as descriptions of royal might or routed foe but as means of bringing these things about. But the twenty-first-century rationalist in me wonders if martial cultures commonly produce such songs or verses simply in response to the desperate need to sustain morale.

Roberts has kept faith with a non-Lankan scholar whose work is a more pure and rarified example of holistic cultural anthropology than any of the Peradeniya crew, Bruce Kapferer, and has bravely followed him in linking the cultural motifs of centuries past to the violence of the 1980s and 90s. Roberts highlights a very real insight of Kapferer’s: the way that threats to the state, to the spiritual realm and to the person can become psychologically conflated. The war poems, with all their gory brutal eloquence on the need to tear apart the enemy, just as the enemy threatens to tear apart the fabric of one’s own world, are profitably read in this light. Roberts goes on to refer to work on the Rwandan massacres, and one could in fact extend this line of thought to consider the St. Bartholomew’s day massacre in France of 1572. Indeed a proper comparison of the latter with the anti-Tamil pogrom of 1983 would surely be instructive, given that in both cases violence spread with astonishing rapidity through the regions, as, in town after town, a civilian majority turned to butcher a minority who had the day before been close neighbours. In both, the excessive use of mutilation and dismemberment presents a clue to the way in which the minority were perceived as being somehow threatening to the sacred fate of the community and the secular fate of the body politic. The comparative fruitfulness of this insight however, does raise the question as to how distinctively Sinhalese such dynamics are.

Some things that are not in it:

This is a problem-oriented work, which draws together expertise and redirects it to the most intellectually worthwhile ends. It is, moreover, designed to do a job in a particular debate about identity, and does not make any attempt to provide a comprehensive account of the epoch. It makes little use of Portuguese and Dutch texts, which might have given a sense of how the troublesome outsiders saw the transformation of Sinhala society over the centuries in question. From a cursory appraisal of the Portuguese sources, what we find is a revolution in perceptions of the Sinhalese over the course of the sixteenth century, as vague references to their warlike and timid nature give way to a reputation as the most fierce opponents of the Portuguese anywhere in Asia. The lurid hatreds and pugnacious ethos of the war-poems or parts of the Rajavaliya were experienced in a very concrete way by the Portuguese. They saw it rear up before them in persistent rebellion. There is, in other words, no dislocation between the rhetorical world of the war-poems and the material world of armed conflict. These perceptions from the outside are also good evidence that the Sinhalese had a well-developed consciousness of their own history: ‘As for the character of the Chingalas, they are generally proud, vain and lazy…because of the antiquity of their Kingdom and nation [nação] and the liberty in which they were always brought up.’ This comes from the pen of Fernão de Queirós, a Portuguese Jesuit writing in the 1680s, who went on to make these comments: ‘In obeying their native Kings they have always been various and inconstant, but most stubborn in not admitting foreign domination, and when the Portuguese entered Ceylon…they did not hesitate to submit to any bold rebel, in order to recover their liberty’; and, ‘their greatest occupation is soldiering, and they enjoy peace only as an accident, and war is the custom.’[4]

There is much more to be had from Queirós’ vast ramshackle text, but even these excerpts give strong independent testimony to the themes that Roberts has managed to draw from the Sinhala material. Given that Queirós, with his seventeenth-century vision of spiritual and temporal conflict, has been a major source for nationalist readings of history it is a bit unusual that Roberts has not turned to the old Jesuit himself. On the other hand, the text is so treacherous, so opaque in its construction and biased in its argument, that it may be better that Roberts did not seek to build part of his case upon it.

Roberts uses the flourishing of the KandyanKingdom, which was the chief Sinhalese polity from the 1590s, to provide his principal time-frame, but he acknowledges that the first example of the war poem genre actually appeared in circa 1585 with the Sitavaka Hatana. It was composed, therefore, before the resurrection of the Kandyan Kingdom, and yet many of the most important elements of the genre are there, fully-formed, including a patriotic turn-of phrase. This alerts us to the fact that at least half-a-century of bloody conflict with the Portuguese has already transpired by this time, latterly waged by its most brilliant warrior-king Rajasinha I. I would argue then that the key development of Sinhala sensibility, the new or renewed orientation towards a xenophobic and patriotic passion for reclaiming land from the outsiders, had its origins in the campaigns of the Kotte rebel, Vidiye Bandara in the 1550s, and then in the self-presentation of the Kings of Sitavaka. And all of this is to neglect another side of the story entirely, that is, the perspective from the lowlands. The view from late sixteenth- to early seventeenth-century Kotte-Colombo would be quite different: it would be that of the Lusitanised, the converted and acquiescent, of the third-or fourth generation Portuguese settlers and their mixed-race offspring, who tried to make Lanka blessed as a Christian rather than Buddhist isle. Their story, and the new identity of ‘Sinhala Christian’ they forged, can be glimpsed in the remarkable Kustantinu Hatana, from 1619-20, a poem which assumes the form of patriotic chauvinism in order to undermine it, singing the praises of the rebel-crusher Portuguese Captain-General Constantino de Sá de Nornonha.

What all this tells us is that Sri Lanka’s experience of military struggle in the sixteenth to nineteenth centuries played a significant part in the hardening of a Sinhala ethnic identity. And, serendipitously, the same kind of argument can be found in Victor Lieberman’s recent comparative analysis of Southeast Asiamainland history.[5] Indeed so much of Robert’s work, which can give one the sense of sawing mightily against the grain of recent orthodoxy in (foreign-based) Lankan studies, suddenly acquires a smooth and harmonious action when viewed by the light of current developments in Southeast Asian scholarship. Roberts’ ‘middle period’ corresponds pretty closely to the ‘post-charter era’ analyzed by Lieberman, and time and again we find points of congruence between the two galactic-political Theravada regions. These can best be appreciated by some extensive quotation from Lieberman. On war:  ‘in many communities intensifying warfare, particularly the chaos of the mid- and late- 1700s, strengthened popular identification with the throne and with official culture by generating mythic reminders of communal danger and salvation, by rendering border communities more dependent on the throne for security, by nurturing ethnic stereotypes, and by integrating conscripts from different locales’ (p. 61). Such responses, we are told, were more likely when the enemy could be easily distinguished by external features or symbols, and when local communities could link their cause to an enduring political entity. They gave rise to a ‘soft’ identity, germane more to the elite than the peasants and prone to recede with the waning of the threat.

Lieberman’s work is so profoundly analytical that it provides countless such tools appropriate for dissecting the Lankan situation. He is more alert than Roberts to the particularly early modern conundrum of ethnic solidification twinned with a voraciously cosmopolitan elite, who participated in fashions, languages, religious networks, dynastic or political allegiances that were trans-local or imperial. For example, Robert’s emphasis on the subordination of outsiders skates over the high status accorded to specialist foreign groups in the early modern maritime city. In Kotte, South Indians gave way to Portuguese as the military specialists, and Lusitanised court culture retained a surprisingly powerful appeal to indigenous rulers throughout the sixteenth and seventeenth centuries. This would also help explain why the island boundaries of Lanka seem to undergo a poetic and political reification in the post-Anuradhapura-period, which nevertheless did little to obstruct the imperial rhetoric of boundless sovereignty. For such reasons, and more, Lieberman is averse to describing identity formation in this period as national or even proto-national, but he settles for a term with which Roberts would heartily concur, ‘politicized ethnicity’, and a conceptual thrust that ought to ring more than a few bells: ‘‘a sense of political community not only proved compatible with, but in fact depended on the maintenance of a deeply hierachical social ethic’ (p. 43).

Or try this point made in debate with Gellner and Anderson: ‘Ultimately all Burmans depended on their king, because as principal patron, he not only authorized subordinate hierarchies, but like Mahasammata, transmitted those principles of order inherent in the cosmos. Burmese records and European accounts show that such images, far from being a conceit of royal ideologues, were deeply rooted in popular consciousness’ (p. 202). Lieberman tells us that while early modern elites on the mainland were more interested in promoting their ritual or religious supremacy than extinguishing minority cultures, they oversaw an increasing homogenization of their realms over the second millennium, and particularly during the sixteenth to nineteenth centuries. The capital could be seen as a machine sucking people in from the regions for rajakariya, religious rites, tribute presentation; and sometimes, it was the capital that went walkabout, when the monarch embarked on pilgrimages or ritual tours of the regions. Roberts will be pleased to find that ‘… popular cultural mobilization relied overwhelmingly on oral channels of communication supplemented by manuscripts’ (p. 43).

He would also find that the principles of his ‘tributary overlordship’ are largely taken for granted in Southeast Asian scholarship. Suddenly it is obvious to which species the political animals of Lanka belong, even if their unique subjection to the hostile ecology of sustained European colonization would ensure a somewhat divergent evolutionary path.

It will be clear by now that Lieberman’s almost teleological obsession with historical progression – and the comparative wealth of both primary and secondary material at his disposal – gives his work a diachronic dimension that is generally absent from Robert’s book.  For example, there is a certain timeless quality to Roberts’ account of the tremendous distance between mundane mortal and divine-like king opened up by court ritual in Kandy, whereas Lieberman describes the contemporary sacral royal theatre of Southeast Asian kings such as Narai of Siam (r. 1656-88) as the pinnacle of a mounting trend towards the centralization of power. It may be that Sri Lanka’s second millennium simply was more static than that of Southeast Asia, an extended come-down after the parties of the first millennium, or it may be that a comparative analysis would reveal that it kept pace in interesting ways.[6] At any rate, Roberts will find far more to cheer in Lieberman’s work than to regret.

There are a few trivial errors of a typographical nature: on p. 73, chart three, where we find that several dates from the 1500s have been turned into the 1600s; Chapter 6, footnote 37, where the origin of the Mahavamsa is placed in 500 BC rather than 500 AD; p. 154, where Rajasinha II is found operating in the sixteenth rather than the seventeenth century; p. 214, Appendix D, where the footnote has been mistakenly repeated from the next page. The text is adorned with some fetching line-drawing illustrations, a useful collation of maps, and some intriguing plates – particularly the late eighteenth-century paintings of Dutch officials in prostrate homage before the King of Kandy, from the Rijksmuseum, Amsterdam.

In its concrete examination and dismissal of various generalizations about the nature of group identity before the rise of the nation state, this work stands as an important intervention in one of the most resonant intellectual debates of our time. It is, however, more likely to spur on debate than to foreclose it, a function of the elliptical evidence and the contemporary relevance of the questions put to it. Nevertheless, a lucid forward by Christopher Bayly reveals its place in a confluence of like-minded scholarship that extends well beyond South east Asia. In such company, Roberts could be excused for feeling that after some years of fairly lonely theoretical resistance he has now come in from the cold.