Imagined Communities
Reflections on the Origin and Spread of Nationalism
—-♦-
BENEDICT ANDERSON
Revised Edition
V
VERSO
London • New York
First published by Verso 1983 This edition published by Verso 2006 © Benedict Anderson, 1983, 1991, 2006 new material © Benedict Anderson, 2006
All rights reserved
The moral rights of the author have been asserted
3579 10 8642
Verso
UK: 6 Meard Street, London W1F 0EG USA: 180 Varick Street, New York, NY 10014-4606 www.versobooks.com
Verso is the imprint of New Left Books
ISBN-13: 978-1-84467-086-4 ISBN-10: 1-84467-086-4
British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress
Typeset by Hewer Text UK Ltd, Edinburgh Printed by Quebecor World, Fairfield
Introduction
Perhaps without being much noticed yet, a fundamental transformation in the history of Marxism and Marxist movements is upon us. Its most visible signs are the recent wars between Vietnam, Cambodia and China. These wars are of world-historical importance because they are the first to occur between regimes whose independence and revolutionary credentials are undeniable, and because none of the belligerents has made more than the most perfunctory attempts to justify the bloodshed in terms of a recognizable Marxist theoretical perspective. While it was still just possible to interpret the Sino-Soviet border clashes of 1969, and the Soviet military interventions in Germany (1953), Hungary (1956), Czechoslovakia (1968), and Afghanistan (1980) in terms of - according to taste - 'social imperialism,' 'defending socialism,' etc., no one, I imagine, seriously believes that such vocabularies have much bearing on what has occurred in Indochina.
If the Vietnamese invasion and occupation of Cambodia in December 1978 and January 1979 represented the first large-scale conventional war waged by one revolutionary Marxist regime against another,   China's assault on Vietnam in February rapidly confirmed
1. This formulation is chosen simply to emphasize the scale and the style of the fighting, not to assign blame. To avoid possible misunderstanding, it should be said that the December 1978 invasion grew out of armed clashes between partisans of the
1
IMAGINED COMMUNITIES
the precedent. Only the most trusting would dare wager that in the declining years of this century any significant outbreak of inter-state hostilities will necessarily find the USSR and the PRC - let alone the smaller socialist states — supporting, or fighting on, the same side. Who can be confident that Yugoslavia and Albania will not one day come to blows? Those variegated groups who seek a withdrawal of the Red Army from its encampments in Eastern Europe should remind themselves of the degree to which its overwhelming presence has, since 1945, ruled out armed conflict between the region's Marxist regimes.
Such considerations serve to underline the fact that since World War II every successful revolution has defined itself in national terms — the People's Republic of China, the Socialist Republic of Vietnam, and so forth — and, in so doing, has grounded itself firmly in a territorial and social space inherited from the prerevolu-tionary past. Conversely, the fact that the Soviet Union shares with the United Kingdom of Great Britain and Northern Ireland the rare distinction of refusing nationality in its naming suggests that it is as much the legatee of the prenational dynastic states of the nineteenth century as the precursor of a twenty-first century internationalist order.2
Eric Hobsbawm is perfectly correct in stating that 'Marxist movements and states have tended to become national not only in form but in substance, i.e., nationalist. There is nothing to suggest
two revolutionary movements going back possibly as far as 1971. After April 1977, border raids, initiated by the Cambodians, but quickly followed by the Vietnamese, grew in size and scope, culminating in the major Vietnamese incursion of December 1977. None of these raids, however, aimed at overthrowing enemy regimes or occupying large territories, nor were the numbers of troops involved comparable to those deployed in December 1978. The controversy over the causes of the war is most thoughtfully pursued in: Stephen P. Heder, 'The Kampuchean-Vietnamese Conflict,' in David W. P. Elliott, ed., The Third Indochina Conflict, pp. 21-67; Anthony Barnett, 'Inter-Communist Conflicts and Vietnam,' Bulletin of Concerned Asian Scholars, 11: 4 (October-December 1979), pp. 2-9; and Laura Summers, 'In Matters of War and Socialism Anthony Barnett would Shame and Honour Kampuchea Too Much,' ibid., pp. 10-18.
2. Anyone who has doubts about the UK's claims to such parity with the USSR should ask himself what nationality its name denotes: Great Brito-Irish?
2
INTRODUCTION
3
that this trend will not continue.' Nor is the tendency confined to the socialist world. Almost every year the United Nations admits new members. And many 'old nations,' once thought fully consolidated, find themselves challenged by 'sub'-nationalisms within their borders - nationalisms which, naturally, dream of shedding this sub-ness one happy day. The reality is quite plain: the 'end of the era of nationalism,' so long prophesied, is not remotely in sight. Indeed, nation-ness is the most universally legitimate value in the political life of our time.
But if the facts are clear, their explanation remains a matter of long-standing dispute. Nation, nationality, nationalism — all have proved notoriously difficult to define, let alone to analyse. In contrast to the immense influence that nationalism has exerted on the modern world, plausible theory about it is conspicuously meagre. Hugh Seton-Watson, author of far the best and most comprehensive English-language text on nationalism, and heir to a vast tradition of liberal historiography and social science, sadly observes: 'Thus I am driven to the conclusion that no "scientific definition" of the nation can be devised; yet the phenomenon has existed and exists.'4 Tom Nairn, author of the path-breaking The Break-up of Britain, and heir to the scarcely less vast tradition of Marxist historiography and social science, candidly remarks: 'The theory of nationalism represents Marxism's great historical failure.'5 But even this confession is somewhat misleading, insofar as it can be taken to imply the regrettable outcome of a long, self-conscious search for theoretical clarity. It would be more exact to say that nationalism has proved an uncomfortable anomaly for Marxist theory and, precisely for that reason, has been largely elided, rather than confronted. How else to explain Marx's failure to explicate the crucial adjective in his memorable formulation of 1848: 'The proletariat of each country
3. Eric Hobsbawm, 'Some Reflections on "The Break-up of Britain" ', New Left Review, 105 (September-October 1977), p. 13.
4. See his Nations and States, p. 5. Emphasis added.
5. See his 'The Modern Janus', New Left Review, 94 (November-December 1975), p. 3. This essay is included unchanged in The Break-up of Britain as chapter 9 (pp. 329-63).
3
IMAGINED COMMUNITIES
must, of course, first of all settle matters with its own bourgeoisie'?6 How else to account for the use, for over a century, of the concept 'national bourgeoisie' without any serious attempt to justify theoretically the relevance of the adjective? Why is this segmentation of the bourgeoisie — a world-class insofar as it is defined in terms of the relations of production — theoretically significant?
The aim of this book is to offer some tentative suggestions for a more satisfactory interpretation of the 'anomaly' of nationalism. My sense is that on this topic both Marxist and liberal theory have become etiolated in a late Ptolemaic effort to 'save the phenomena'; and that a reorientation of perspective in, as it were, a Copernican spirit is urgently required. My point of departure is that nationality, or, as one might prefer to put it in view of that word's multiple significations, nation-ness, as well as nationalism, are cultural artefacts of a particular kind. To understand them properly we need to consider carefully how they have come into historical being, in what ways their meanings have changed over time, and why, today, they command such profound emotional legitimacy. I will be trying to argue that the creation of these artefacts towards the end of the eighteenth century was the spontaneous distillation of a complex 'crossing' of discrete historical forces; but that, once created, they became 'modular,' capable of being transplanted, with varying degrees of self-consciousness, to a great variety of social terrains, to merge and be merged with a correspondingly wide variety of political and ideological constellations. I will also attempt to show why these particular cultural artefacts have aroused such deep attachments.
6. Karl Marx and Friedrich Engels, The Communist Manifesto, in the Selected Works, I, p. 45. Emphasis added. In any theoretical exegesis, the words 'of course' should flash red lights before the transported reader.
7. As Aira Kemilainen notes, the twin 'founding fathers' of academic scholarship on nationalism, Hans Kohn and Carleton Hayes, argued persuasively for this dating. Their conclusions have, I think, not been seriously disputed except by nationalist ideologues in particular countries. Kemilainen also observes that the word 'nationalism' did not come into wide general use until the end of the nineteenth century. It did not occur, for example, in many standard nineteenth century lexicons. If Adam Smith conjured with the wealth of'nations,' he meant by the term no more than 'societies' or 'states.' Aira Kemilainen, Nationalism, pp. 10, 33, and 48—49.
4
INTRODUCTION
CONCEPTS AND DEFINITIONS
Before addressing the questions raised above, it seems advisable to consider briefly the concept of'nation' and offer a workable definition. Theorists of nationalism have often been perplexed, not to say irritated, by these three paradoxes: (1) The objective modernity of nations to the historian's eye vs. their subjective antiquity in the eyes of nationalists. (2) The formal universality of nationality as a socio-cultural concept — in the modern world everyone can, should, will 'have' a nationality, as he or she 'has' a gender — vs. the irremediable particularity of its concrete manifestations, such that, by definition, 'Greek' nationality is sui generis. (3) The 'political' power of nationalisms vs. their philosophical poverty and even incoherence. In other words, unlike most other isms, nationalism has never produced its own grand thinkers: no Hobbeses, Tocquevilles, Marxes, or Webers. This 'emptiness' easily gives rise, among cosmopolitan and polylingual intellectuals, to a certain condescension. Like Gertrude Stein in the face of Oakland, one can rather quickly conclude that there is 'no there there'. It is characteristic that even so sympathetic a student of nationalism as Tom Nairn can nonetheless write that: ' "Nationalism" is the pathology of modern developmental history, as inescapable as "neurosis" in the individual, with much the same essential ambiguity attaching to it, a similar built-in capacity for descent into dementia, rooted in the dilemmas of helplessness thrust upon most of the world (the equivalent of infantilism for societies) and largely incurable.'8
Part of the difficulty is that one tends unconsciously to hypos-tasize the existence of Nationalism-with-a-big-N (rather as one might Age-with-a-capital-A) and then to classify 'it' as an ideology. (Note that if everyone has an age, Age is merely an analytical expression.) It would, I think, make things easier if one treated it as if it belonged with 'kinship' and 'religion', rather than with 'liberalism' or 'fascism'.
In  an  anthropological  spirit,  then,   I  propose  the following
8. The Break-up of Britain, p. 359.
5
IMAGINED COMMUNITIES
definition of the nation: it is an imagined political community - and imagined as both inherently limited and sovereign.
It is imagined because the members of even the smallest nation will never know most of their fellow-members, meet them, or even hear of them, yet in the minds of each lives the image of their communion.9 Renan referred to this imagining in his suavely back-handed way when he wrote that 'Or l'essence d'une nation est que tous les individus aient beaucoup de choses en commun, et aussi que tous aient oublie bien des choses.' With a certain ferocity Gellner makes a comparable point when he rules that 'Nationalism is not the awakening of nations to self-consciousness: it invents nations where they do not exist.' The drawback to this formulation, however, is that Gellner is so anxious to show that nationalism masquerades under false pretences that he assimilates 'invention' to 'fabrication' and 'falsity', rather than to 'imagining' and 'creation'. In this way he implies that 'true' communities exist which can be advantageously juxtaposed to nations. In fact, all communities larger than primordial villages of face-to-face contact (and perhaps even these) are imagined. Communities are to be distinguished, not by their falsity/genuineness, but by the style in which they are imagined. Javanese villagers have always known that they are connected to people they have never seen, but these ties were once imagined particularistically - as indefinitely stretchable nets of kinship and clientship. Until quite recently, the Javanese language had no word meaning the abstraction 'society.' We may today think of the French aristocracy of the ancien regime as a class; but surely it was
9. Cf. Seton-Watson, Nations and States, p. 5: 'All that I can find to say is that a nation exists when a significant number of people in a community consider themselves to form a nation, or behave as if they formed one.' We may translate 'consider themselves' as 'imagine themselves.'
10. Ernest Renan, 'Qu'est-ce qu'une nation?' in OEuvres Completes, 1, p. 892. He adds: 'tout citoyen francais doit avoir oublie la Saint-Barthelemy, les massacres du Midi an XHIe siecle. II n'y a pas en France dix families qui puissent fournir la preuve d'une origine franque . . .'
11. Ernest Gellner, Thought and Change, p. 169. Emphasis added.
6
INTRODUCTION
imagined this way only very late. To the question 'Who is the Comte de X?' the normal answer would have been, not 'a member of the aristocracy,' but 'the lord of X,' 'the uncle of the Baronne de Y,' or 'a client of the Due de Z.'
The nation is imagined as limited because even the largest of them, encompassing perhaps a billion living human beings, has finite, if elastic, boundaries, beyond which lie other nations. No nation imagines itself coterminous with mankind. The most messianic nationalists do not dream of a day when all the members of the human race will join their nation in the way that it was possible, in certain epochs, for, say, Christians to dream of a wholly Christian planet.
It is imagined as sovereign because the concept was born in an age in which Enlightenment and Revolution were destroying the legitimacy of the divinely-ordained, hierarchical dynastic realm. Coming to maturity at a stage of human history when even the most devout adherents of any universal religion were inescapably confronted with the living pluralism of such religions, and the allomorphism between each faith's ontological claims and territorial stretch, nations dream of being free, and, if under God, directly so. The gage and emblem of this freedom is the sovereign state.
Finally, it is imagined as a community, because, regardless of the actual inequality and exploitation that may prevail in each, the nation is always conceived as a deep, horizontal comradeship. Ultimately it is this fraternity that makes it possible, over the past two centuries, for so many millions of people, not so much to kill, as willingly to die for such limited imaginings.
These deaths bring us abruptly face to face with the central problem posed by nationalism: what makes the shrunken imaginings of recent history (scarcely more than two centuries) generate such colossal sacrifices? I believe that the beginnings of an answer lie in the cultural roots of nationalism.
12. Hobsbawm, for example, 'fixes' it by saying that in 1789 it numbered about 400,000 in a population of23,000,000. (See his The Age of Revolution, p. 78). But would this statistical picture of the noblesse have been imaginable under the ancien regime?
7
BLANK PAGE
===== 2 =
Cultural Roots
No more arresting emblems of the modern culture of nationalism exist than cenotaphs and tombs of Unknown Soldiers. The public ceremonial reverence accorded these monuments precisely because they are either deliberately empty or no one knows who lies inside them, has no true precedents in earlier times. To feel the force of this modernity one has only to imagine the general reaction to the busy-body who 'discovered' the Unknown Soldier's name or insisted on filling the cenotaph with some real bones. Sacrilege of a strange, contemporary kind! Yet void as these tombs are of identifiable mortal remains or immortal souls, they are nonetheless saturated with ghostly national imaginings. (This is why so many different nations have such
1. The ancient Greeks had cenotaphs, but for specific, known individuals whose bodies, for one reason or another, could not be retrieved for regular burial. I owe this information to my Byzantinist colleague Judith Herrin.
2. Consider, for example, these remarkable tropes: 1. 'The long grey line has never failed us. Were you to do so, a million ghosts in olive drab, in brown khaki, in blue and grey, would rise from their white crosses, thundering those magic words: Duty, honour, country.' 2. 'My estimate of [the American man-at-arms] was formed on the battlefield many, many years ago, and has never changed. I regarded him then, as I regard him now, as one of the world's noblest figures; not only as one of the finest military characters, but also as one of the most stainless [sic]. . ....'He belongs to history as
furnishing one of the greatest examples of successful patriotism [sic]. He belongs to posterity as the instructor of future generations in the principles of liberty and freedom.
9
IMAGINED COMMUNITIES
tombs without feeling any need to specify the nationality of their absent occupants. What else could they be but Germans, Americans, Argentinians . . .?)
The cultural significance of such monuments becomes even clearer if one tries to imagine, say, a Tomb of the Unknown Marxist or a cenotaph for fallen Liberals. Is a sense of absurdity avoidable? The reason is that neither Marxism nor Liberalism is much concerned with death and immortality. If the nationalist imagining is so concerned, this suggests a strong affinity with religious imaginings. As this affinity is by no means fortuitous, it may be useful to begin a consideration of the cultural roots of nationalism with death, as the last of a whole gamut of fatalities.
If the manner of a man's dying usually seems arbitrary, his mortality is inescapable. Human fives are full of such combinations of necessity and chance. We are all aware of the contingency and ineluctability of our particular genetic heritage, our gender, our life-era, our physical capabilities, our mother-tongue, and so forth. The great merit of traditional religious world-views (which naturally must be distinguished from their role in the legitimation of specific systems of domination and exploitation) has been their concern with man-in-the-cosmos, man as species being, and the contingency of life. The extraordinary survival over thousands of years of Buddhism, Christianity or Islam in dozens of different social formations attests to their imaginative response to the overwhelming burden of human suffering — disease, mutilation, grief, age, and death. Why was I born blind? Why is my best friend paralysed? Why is my daughter retarded? The religions attempt to explain. The great weakness of all evolutionary/progressive styles of thought, not excluding Marx-ism, is that such questions are answered with impatient silence. At
He belongs to the present, to us, by his virtues and his achievements.' Douglas MacArthur, 'Duty, Honour, Country,' Address to the U.S. Military Academy, West Point, May 12, 1962, in his A Soldier Speaks, pp. 354 and 357.
3. Cf. Regis Debray, 'Marxism and the National Question,' New Left Review, 105 (September-October 1977), p. 29. In the course of doing fieldwork in Indonesia in the 1960s I was struck by the calm refusal of many Muslims to accept the ideas of Darwin. At first I interpreted this refusal as obscurantism. Subsequently I came to see it as an honourable attempt to be consistent: the doctrine of evolution was simply not compatible with the teachings of Islam. What are we to make of a scientific materialism
10
CULTURAL ROOTS
the same time, in different ways, religious thought also responds to obscure intimations of immortality, generally by transforming fatality into continuity (karma, original sin, etc.). In this way, it concerns itself with the links between the dead and the yet unborn, the mystery of re-generation. Who experiences their child's conception and birth without dimly apprehending a combined connectedness, fortuity, and fatality in a language of 'continuity'? (Again, the disadvantage of evolutionary/progressive thought is an almost Her-acHtean hostility to any idea of continuity.)
I bring up these perhaps simpleminded observations primarily because in Western Europe the eighteenth century marks not only the dawn of the age of nationalism but the dusk of religious modes of thought. The century of the Enlightenment, of rationalist secularism, brought with it its own modern darkness. With the ebbing of religious belief, the suffering which belief in part composed did not disappear. Disintegration of paradise: nothing makes fatality more arbitrary. Absurdity of salvation: nothing makes another style of continuity more necessary. What then was required was a secular transformation of fatality into continuity, contingency into meaning. As we shall see, few things were (are) better suited to this end than an idea of nation. If nation-states are widely conceded to be 'new' and 'historical,' the nations to which they give political expression always loom out of an immemorial past,4 and, still more important, glide
which formally accepts the findings of physics about matter, yet makes so little effort to link these findings with the class struggle, revolution, or whatever. Does not the abyss between protons and the proletariat conceal an unacknowledged metaphysical conception of man? But see the refreshing texts of Sebastiano Timpanaro, On Materialism and The Freudian Slip, and Raymond Williams' thoughtful response to them in 'Timpanaro's Materialist Challenge,' New Left Review, 109 (May-June 1978), pp. 3-17.
4. The late President Sukarno always spoke with complete sincerity of the 350 years of colonialism that his 'Indonesia' had endured, although the very concept 'Indonesia' is a twentieth-century invention, and most of today's Indonesia was only conquered by the Dutch between 1850 and 1910. Preeminent among contemporary Indonesia's national heroes is the early nineteenth-century Javanese Prince Diponegoro, although the Prince's own memoirs show that he intended to 'conquer [not liberate!] Java,' rather than expel 'the Dutch.' Indeed, he clearly had no concept of'the Dutch' as a collectivity. See Harry J. Benda and John A. Larkin, eds., The World of Southeast Asia, p. 158; and Ann Kumar, 'Diponegoro (1778?-1855),' Indonesia, 13 (April 1972), p. 103.
11
IMAGINED COMMUNITIES
into a limitless future. It is the magic of nationalism to turn chance into destiny. With Debray we might say, 'Yes, it is quite accidental that I am born French; but after all, France is eternal.'
Needless to say, I am not claiming that the appearance of nationalism towards the end of the eighteenth century was 'produced' by the erosion of religious certainties, or that this erosion does not itself require a complex explanation. Nor am I suggesting that somehow nationalism historically 'supersedes' religion. What I am proposing is that nationalism has to be understood by aligning it, not with selfconsciously held political ideologies, but with the large cultural systems that preceded it, out of which — as well as against which - it came into being.
For present purposes, the two relevant cultural systems are the religious community and the dynastic realm. For both of these, in their heydays, were taken-for-granted frames of reference, very much as nationality is today. It is therefore essential to consider what gave these cultural systems their self-evident plausibility, and at the same time to underline certain key elements in their decomposition.
THE RELIGIOUS COMMUNITY
Few things are more impressive than the vast territorial stretch of the Ummah Islam from Morocco to the Sulu Archipelago, of Christendom from Paraguay to Japan, and of the Buddhist world from Sri Lanka to the Korean peninsula. The great sacral cultures (and for our purposes here it may be permissible to include 'Confucianism') incorporated conceptions of immense communities. But Christendom, the Islamic Ummah, and even the Middle Kingdom — which, though we think of it today as Chinese, imagined itself not as Chinese, but as
Emphasis added. Similarly, Kemal Atatiirk named one of his state banks the Eti Banka (Hittite Bank) and another the Sumerian Bank. (Seton-Watson, Nations and States, p. 259). These banks flourish today, and there is no reason to doubt that many Turks, possibly not excluding Kemal himself, seriously saw, and see, in the Hittites and Sumerians their Turkish forebears. Before laughing too hard, we should remaind ourselves of Arthur and Boadicea, and ponder the commercial success of Tolkien's mythographies.
12
CULTURAL ROOTS
central - were imaginable largely through the medium of a sacred language and written script. Take only the example of Islam: if Maguindanao met Berbers in Mecca, knowing nothing of each other's languages, incapable of communicating orally, they nonetheless understood each other's ideographs, because the sacred texts they shared existed only in classical Arabic. In this sense, written Arabic functioned like Chinese characters to create a community out of signs, not sounds. (So today mathematical language continues an old tradition. Of what the Thai call + Rumanians have no idea, and vice versa, but both comprehend the symbol.) All the great classical communities conceived of themselves as cosmically central, through the medium of a sacred language linked-to. a superterrestrial order of power. Accordingly, the stretch of written Latin, Pali, Arabic, or Chinese was, in theory, unlimited. (In fact, the deader the written language - the farther it was from speech — the better: in principle everyone has access to a pure world of signs.)
Yet such classical communities linked by sacred languages had a character distinct from the imagined communities of modern nations. One crucial difference was the older communities' confidence in the unique sacredness of their languages, and thus their ideas about admission to membership. Chinese mandarins looked with approval on barbarians who painfully learned to paint Middle Kingdom ideograms. These barbarians were already halfway to full absorption. Half-civilized was vastly better than barbarian. Such an attitude was certainly not peculiar to the Chinese, nor confined to antiquity. Consider, for example, the following 'policy on barbarians' formulated by the early-nineteenth-century Colombian liberal Pedro Fermin de Vargas:
To expand our agriculture it would be necessary to hispanicize our Indians. Their idleness, stupidity, and indifference towards normal endeavours causes one to think that they come from a degenerate race which deteriorates in proportion to the distance from its origin . . . it would be very desirable that the Indians be extinguished, by miscegenation with
5. Hence the equanimity with which Sinicized Mongols and Manchus were accepted as Sons of Heaven.
13
IMAGINED COMMUNITIES
the whites, declaring them free of tribute and other charges, and giving them private property in land.6
How striking it is that this liberal still proposes to 'extinguish' his Indians in part by 'declaring them free of tribute' and 'giving them private property in land', rather than exterminating them by gun and microbe as his heirs in Brazil, Argentina, and the United States began to do soon afterwards. Note also, alongside the condescending cruelty, a cosmic optimism: the Indian is ultimately redeemable — by impregnation with white, 'civilized' semen, and the acquisition of private property, like everyone else. (How different Fermin's attitude is from the later European imperialist's preference for 'genuine' Malays, Gurkhas, and Hausas over 'half-breeds,' 'semi-educated natives,' 'wogs', and the like.)
Yet if the sacred silent languages were the media through which the great global communities of the past were imagined, the reality of such apparitions depended on an idea largely foreign to the contemporary Western mind: the non-arbitrariness of the sign. The ideograms of Chinese, Latin, or Arabic were emanations of reality, not randomly fabricated representations of it. We are familiar with the long dispute over the appropriate language (Latin or vernacular) for the mass. In the Islamic tradition, until quite recently, the Qur'an was literally untranslatable (and therefore untranslated), because Allah's truth was accessible only through the unsubstitutable true signs of written Arabic. There is no idea here of a world so separated from language that all languages are equidistant (and thus interchangeable) signs for it. In effect, ontological reality is apprehensible only through a single, privileged system of re-presentation: the truth-language of Church Latin, Qur'anic Arabic, or Examination Chinese. And, as truth-languages, imbued with an impulse largely foreign to
6. John Lynch, The Spanish-American Revolutions, 1808—1826, p. 260. Emphasis added.
7. Church Greek seems not to have achieved the status of a truth-language. The reasons for this 'failure' are various, but one key factor was certainly the fact that Greek remained a living demotic speech (unlike Latin) in much of the Eastern Empire. This insight I owe to Judith Herrin.
14
CULTURAL ROOTS
nationalism, the impulse towards conversion. By conversion, I mean not so much the acceptance of particular religious tenets, but alchemic absorption. The barbarian becomes 'Middle Kingdom', the Rif Muslim, the Ilongo Christian. The whole nature of man's being is sacrally malleable. (Contrast thus the prestige of these old world-languages, towering high over all vernaculars, with Esperanto or Volapiik, which lie ignored between them.) It was, after all, this possibility of conversion through the sacred language that made it possible for an 'Englishman' to become Pope8 and a 'Manchu' Son of Heaven.
But even though the sacred languages made such communities as Christendom imaginable, the actual scope and plausibility of these communities can not be explained by sacred script alone: their readers were, after all, tiny literate reefs on top of vast illiterate oceans. A fuller explanation requires a glance at the relationship between the literati and their societies. It would be a mistake to view the former as a kind of theological technocracy. The languages they sustained, if abstruse, had none of the self-arranged abstruseness of lawyers' or economists' jargons, on the margin of society's idea of reality. Rather, the literati were adepts, strategic strata in a cosmological hierarchy of which the apex was divine. The fundamental conceptions about 'social groups' were centripetal and hierarchical, rather than boundary-oriented and horizontal. The astonishing power of the papacy in its noonday is only comprehensible in terms of a trans-European Latin-writing clerisy, and sl conception of the world, shared by virtually everyone, that the bilingual intelligentsia, by mediating between vernacular and Latin, mediated
8. Nicholas Brakespear held the office of pontiff between 1154 and 1159 under the name Adrian IV.
9. Marc Bloch reminds us that 'the majority of lords and many great barons [in mediaeval times] were administrators incapable of studying personally a report or an account.' Feudal Society, I, p. 81.
10. This is not to say that the illiterate did not read. What they read, however, was not words but the visible world. 'In the eyes of all who were capable of reflection the material world was scarcely more than a sort of mask, behind which took place all the really important things; it seemed to them also a language, intended to express by signs a more profound reality.' Ibid. p. 83.
15
IMAGINED COMMUNITIES
between earth and heaven. (The awesomeness of excommunication reflects this cosmology.)
Yet for all the grandeur and power of the great religiously imagined communities, their unselfconscious coherence waned steadily after the late Middle Ages. Among the reasons for this decline, I wish here to emphasize only the two which are directly related to these communities' unique sacredness.
First was the effect of the explorations of the non-European world, which mainly but by no means exclusively in Europe 'abruptly widened the cultural and geographic horizon and hence also men's conception of
11
possible forms of human life.' The process is already apparent in the greatest of all European travel-books. Consider the following awed description of Kublai Khan by the good Venetian Christian Marco Polo
12
at the end of the thirteenth century:
The grand khan, having obtained this signal victory, returned with great pomp and triumph to the capital city of Kanbalu. This took place in the month of November, and he continued to reside there during the months of February and March, in which latter was our festival of Easter. Being aware that this was one of our principal solemnities, he commanded all the Christians to attend him, and to bring with them their Book, which contains the four Gospels of the Evangelists. After causing it to be repeatedly perfumed with incense, in a ceremonious manner, he devoutly kissed it, and directed that the same should be done by all his nobles who were present. This was his usual practice upon each of the principal Christian festivals, such as Easter and Christmas; and he observed the same at the festivals of the Saracens, Jews, and idolaters. Upon being asked his motive for this conduct, he said: 'There are four great Prophets who are reverenced and worshipped by the different classes of mankind. The Christians regard Jesus Christ as their divinity; the Saracens, Mahomet; the Jews, Moses; and the idolaters, Sogomombar-kan, the most eminent among their idols. I do honour and show respect to all the four,
11. Erich Auerbach, Mimesis, p. 282.
12. Marco Polo, The Travels of Marco Polo, pp. 158-59. Emphases added. Notice that, though kissed, the Evangel is not read.
16
CULTURAL ROOTS
and invoke to my aid whichever amongst them is in truth supreme in heaven.' But from the manner in which his majesty acted towards them, it is evident that he regarded the faith of the Christians as the truest and the best ...
What is so remarkable about this passage is not so much the great Mongol dynast's calm religious relativism (it is still a religious relativism), as Marco Polo's attitude and language. It never occurs to him, even though he is writing for fellow-European Christians, to term Kublai a hypocrite or an idolater. (No doubt in part because 'in respect to number of subjects, extent of territory, and amount of revenue, he surpasses every sovereign that has heretofore been or that now is in the world.') And in the unselfconscious use of'our' (which becomes 'their'), and the description of the faith of the Christians as 'truest,' rather than 'true,' we can detect the seeds of a territorialization of faiths which foreshadows the language of many nationalists ('our' nation is 'the best' — in a competitive, comparative field).
What a revealing contrast is provided by the opening of the letter written by the Persian traveller 'Rica' to his friend 'Ibben' from Paris in '1712':14
The Pope is the chief of the Christians; he is an ancient idol, worshipped now from habit. Once he was formidable even to princes, for he would depose them as easily as our magnificent sultans depose the kings of Iremetia or Georgia. But nobody fears him any longer. He claims to be the successor of one of the earliest Christians, called Saint Peter, and it is certainly a rich succession, for his treasure is immense and he has a great country under his control.
The deliberate, sophisticated fabrications of the eighteenth century Catholic mirror the naive realism of his thirteenth-century predecessor, but by now the 'relativization' and 'territorialization' are utterly self-conscious, and political in intent. Is it unreasonable to see a paradoxical
13. The Travels of Marco Polo, p. 152.
14. Henri de Montesquieu, Persian Letters, p. 81. The Lettres Persanes first appeared in 1721.
17
IMAGINED COMMUNITIES
elaboration of this evolving tradition in the Ayatollah Ruhollah Khomeini's identification of The Great Satan, not as a heresy, nor even as a demonic personage (dim litde Carter scarcely fitted the bill), but as a nation?
Second was a gradual demotion of the sacred language itself. Writing of mediaeval Western Europe, Bloch noted that 'Latin was not only the language in which teaching was done, it was the only language taught.'15 (This second 'only' shows quite clearly the sacredness of Latin - no other language was thought worth the teaching.) But by the sixteenth century all this was changing fast. The reasons for the change need not detain us here: the central importance of print-capitalism will be discussed below. It is sufficient to remind ourselves of its scale and pace. Febvre and Martin estimate that 77% of the books printed before 1500 were still in Latin
16
(meaning nonetheless that 23% were already in vernaculars). If of the 88 editions printed in Paris in 1501 all but 8 were in Latin, after 1575 a
17
majority were always in French. Despite a temporary come-back during the Counter-Reformation, Latin's hegemony was doomed. Nor are we speaking simply of a general popularity. Somewhat later, but at no less dizzying speed, Latin ceased to be the language of a pan-European high intelligentsia. In the seventeenth century Hobbes (1588-1678) was a figure of continental renown because he wrote in the truth-language. Shakespeare (1564—1616), on the other hand, composing in the
18
vernacular, was virtually unknown across the Channel. And had English not become, two hundred years later, the pre-eminent world-imperial language, might he not largely have retained his original insular obscurity? Meanwhile, these men's cross-Channel near-contemporaries, Descartes (1596-1650) and Pascal (1623-1662), conducted most of their correspondence in Latin; but virtually all of Voltaire's (1694—
19
1778) was in the vernacular. 'After 1640, with fewer and fewer books coming out in Latin, and more and more in the vernacular languages, publishing was ceasing to be an international [sic]
15. Bloch, Feudal Society, I, p. 77. Emphasis added.
16. Lucien Febvre and Henri-Jean Martin, The Coming of the Book, pp. 248—49.
17. Ibid., p. 321.
18. Ibid., p. 330.
19. Ibid., pp. 331-32.
18
CULTURAL ROOTS
enterprise.' In a word, the fall of Latin exemplified a larger process in which the sacred communities integrated by old sacred languages were gradually fragmented, pluralized, and territorialized.
THE DYNASTIC REALM
These days it is perhaps difficult to put oneself empathetically into a world in which the dynastic realm appeared for most men as the only imaginable 'political' system. For in fundamental ways 'serious' monarchy lies transverse to all modern conceptions of political life. Kingship organizes everything around a high centre. Its legitimacy derives from divinity, not from populations, who, after all, are subjects, not citizens. In the modern conception, state sovereignty is fully, flatly, and evenly operative over each square centimetre of a legally demarcated territory. But in the older imagining, where states were defined by centres, borders were porous and indistinct, and sovereignties faded imperceptibly into one another. Hence, paradoxically enough, the ease with which pre-modern empires and kingdoms were able to sustain their rule over immensely heterogeneous, and often not even contiguous, populations for long periods of
22
time.
One must also remember that these antique monarchical states
20. Ibid., pp. 232-33. The original French is more modest and historically exact: 'Tandis que Ton edite de moins en moins d'ouvrages en latin, et une proportion toujours plus grande de textes en langue nationale, le commerce du livre se morcelle en Europe.' L'Apparition du Livre, p. 356.
21. Notice the displacement in rulers' nomenclature that corresponds to this transformation. Schoolchildren remember monarchs by their first names (what was William the Conqueror's surname?), presidents by their last (what was Ebert's Christian name?). In a world of citizens, all of whom are theoretically eligible for the presidency, the limited pool of 'Christian' names makes them inadequate as specifying designators. In monarchies, however, where rule is reserved for a single surname, it is necessarily 'Christian' names, with numbers, or sobriquets, that supply the requisite distinctions.
22. We may here note in passing that Nairn is certainly correct in describing the 1707 Act of Union between England and Scotland as a 'patrician bargain,' in the sense that the union's architects were aristocratic politicians. (See his lucid discussion in The Break-up of Britain, pp. 136f). Still, it is difficult to imagine such a bargain being
19
IMAGINED COMMUNITIES
expanded not only by warfare but by sexual politics — of a kind very different from that practised today. Through the general principle of verticality, dynastic marriages brought together diverse populations under new apices. Paradigmatic in this respect was the House of Habsburg. As the tag went, Bella gerant alii, tufelix Austria nube! Here, in somewhat abbreviated form, is the later dynasts' titulature.
Emperor of Austria; King of Hungary, of Bohemia, of Dalmatia, Croatia, Slavonia, Galicia, Lodomeria, and Illyria; King of Jerusalem, etc; Archduke of Austria [sic]; Grand Duke of Tuscany and Cracow; Duke of Loth[a]ringia, of Salzburg, Styria, Carinthia, Carniola, and Bukovina; Grand Duke of Transylvania, Margrave of Moravia; Duke of Upper and Lower Silesia, of Modena, Parma, Piacenza, and Guastella, of Ausschwitz and Sator, of Teschen, Friaul, Ragusa, and Zara; Princely Count of Habsburg and Tyrol, of Kyburg, Gorz, and Gradiska; Duke of Trient and Brizen; Margrave of Upper and Lower Lausitz and in Istria; Count of Hohenembs, Feldkirch, Bregenz, Sonnenberg, etc.; Lord of Trieste, of Cattaro, and above the Windisch Mark; Great Voyvod of the Voyvodina, Servia .... etc.
This, Jaszi justly observes, was, 'not without a certain comic aspect . . . the record of the innumerable marriages, hucksterings and captures of the Habsburgs.'
In realms where polygyny was religiously sanctioned, complex systems of tiered concubinage were essential to the integration of the realm. In fact, royal lineages often derived their prestige, aside from any aura of divinity, from, shall we say, miscegenation?   For such
struck between the aristocracies of two republics. The conception of a United Kingdom was surely the crucial mediating element that made the deal possible.
23. Oscar Jaszi, The Dissolution of the Habsburg Monarchy, p. 34.
24. Most notably in pre-modern Asia. But the same principle was at work in monogamous Christian Europe. In 1910, one Otto Forst put out his Ahnentafel Seiner Kaiserlichen und Königlichen Hoheit des durchlauchtigsten Hem Erzherzogs Franz Ferdinand, listing 2,047 of the soon-to-be-assassinated Archduke's ancestors. They included 1,486 Germans, 124 French, 196 Italians, 89 Spaniards, 52 Poles, 47 Danes, 20 Englishmen/women, as well as four other nationalities. This 'curious document' is cited in ibid., p. 136, no. 1. I can not resist quoting here Franz Joseph's wonderful
20