Chapter 7

Reimagined Communities
Benedict Anderson’s Theory of Nationalism1

Introduction

“Proverbially,” writes Benedict Anderson in a new afterword to Imagined Communities, “a writer loses his/her book at the moment that it is published and enters the public sphere.”2 Anderson is thinking specifically about the difficulties experienced by authors when their work is translated into languages they do not understand. Given that his book has been translated into twenty-seven since it first appeared in 1983, these are difficulties with which even the multi-lingual Anderson is now overly familiar. Yet the point is of more general application. For if there is one book on nationalism that every student is expected to read, one book that is certain to be included in any survey of the competing theories, it is Imagined Communities. It is, as Josep Llobera has written of the core conception, “as if people had been waiting for such an expression to be coined.”3 That expression can now be encountered in quite unexpected places. The following passage, chosen virtually at random, is from a discussion by Alasdair MacIntyre of the possibility that poetry might also embody a political philosophy:

One cannot be a monk or a member of café society without being able to imagine oneself as such; living the role of either is a form of imaginative acting out. And so too with the community of a nation: to be Irish or English, I must be able to imagine myself as Irish or English, something achievable in part by participating in the shared poetic utterance of the nation. Take away shared songs and poetry, shared monuments and architecture, shared imaginative conceptions of what is for this nation sacred ground and you at the very least weaken the bonds of nationality. So nations to be real must first be imagined.4

MacIntyre does not refer to Anderson here, but this is not inadequate scholarship: he does not need to. The concept of imagined community has become so much a part of contemporary intellectual common sense that citation is virtually superfluous. And MacIntyre at least understands one of the points Anderson was trying to make. All too often the title has been quoted with such little comprehension that Anderson himself can now describe it as “a pair of words from which the vampires of banality have by now sucked almost all the blood.”5

The appearance of the third edition therefore gives us an opportunity to reassess this original and influential work, but also to identify some of the problems to which it gives rise, problems that are, in part, the very reason for its popularity. These are not all simply the result of the inevitable misunderstandings that occur when a complex concept is seized upon to fill an explanatory gap; when, as Philip Spencer and Howard Wollman put it, “invocation” becomes “a substitute for analysis.”6 Some are the result of Benedict Anderson’s underlying theoretical assumptions. As he himself notes, “the book attempted to combine a kind of historical materialism with what later on came to be called discourse analysis. Marxist modernism married to post-modernism avant la lettre.”7 It is the postmodern aspects of the work that have proved the most influential, all too often at the expense of Anderson’s Marxism. Nevertheless, it would be ungenerous not to begin by recognizing his achievement. To understand why this book had such an impact, it is necessary first to review how nationalism had previously been dealt with in the Marxist tradition.

Strategies, definitions, explanations

Marx and Engels engaged with the issue of nationalism in the middle decades of the nineteenth century—in other words, during the period in which the bourgeois revolution was being completed across Western Europe, North America, and Japan. They argued that the working class (and “the democracy” more generally) should support national movements and the formation of new nation-states where they would hasten the development of capitalism and consequently the emergence of a working class, and where they would weaken the great reactionary powers of Europe, the most powerful of which was absolutist Russia.8

Self-determination was not necessarily the absolute priority. Marx and Engels rejected the view that every national group had the right to establish a state, the so-called principle of nationality, as it was then known.9 On the contrary, for them it entirely depended on whether the success of the movement was likely to lead to a progressive outcome or not. Nor was their attitude to a particular movement determined by the class nature or political attitudes of its leadership. The Hungarian rising of 1848 was dominated by the nobility, the aristocracy led the Polish insurrection of 1863, and even the Irish Fenians—in many respects one of the more politically advanced nonsocialist groups of the time—were heavily influenced by the Catholic Church. None of these negative characteristics was decisive, however, compared with the positive objective consequences of opening up the possibilities for capitalist development or closing down the influence of the absolutist states. By contrast, Marx and Engels refused support to the Czechs and southern Slavs during the revolutions of 1848–49 because they were backed by Russian absolutism—the “gendarme of Europe”—for its own purposes.10

The specific situations with which these socialist strategies toward nationalism were intended to deal are now largely historical, but the method employed remains of enduring value. It was, however, arrived at without any real explanation of the emergence or nature of nations, whose existence Marx and Engels essentially took for granted. They left the movement a correct strategic orientation on national movements together with an undeveloped theoretical position on the nature of nations, national consciousness, and so on.

The next generation of Marxists quite understandably concentrated, as a matter of practical necessity, on refining the approach of revolutionary socialists to national movements and national demands under the changed conditions of the imperialist era. These discussions, which extended from the mid-1890s to the debates on “the national and colonial question” during the first four congresses of the Communist International (1919–1922), represent one of Marxism’s greatest contributions to the question of socialist strategy.11 But although some participants, notably Kautsky and Lenin, made attempts to explain how nations emerged, these were rarely central to the argument and usually went no further than emphasizing the need for capitalism to dominate a territorial home market and the role of language in unifying the inhabitants of that territory.12

The Austro-Marxist tendency, represented by Karl Renner and Otto Bauer, did focus on the question of national formation, above all in Bauer’s monumental The National Question and Social Democracy (1906). But the definition of a nation offered by Bauer was resolutely nonmaterialist: “The nation is the totality of human beings bound together by a community of fate into a community of character.”13 Bauer did see capitalism as playing a role in the development of national consciousness, but only in the sense that such consciousness can only be complete when it is aware of other nations and the difference between them, which occurs most fully under capitalist development. Bauer’s work has been hailed as the only serious Marxist attempt to deal with the national question, but mainly by people who welcome it precisely because of its distance from Marxism. (Indeed, even Bauer is too marked by “economism” and “class reductionism” for some of his present day admirers.14)

Lenin claimed that Bauer’s theory was “basically psychological” and endorsed instead the “historico-economic” explanation associated with Kautsky and in his own writings, but failed to propose a comparably detailed alternative explanation for the emergence of nations or the nature of national consciousness.15 What was offered in direct opposition to Bauer was not a counterexplanation but a counterdefinition that is unfortunately still widely accepted by many on the left today. In 1913 Joseph Stalin wrote, under Lenin’s guidance: “A nation is a historically constituted, stable community of people, formed on the basis of a common language, territory, economic life, and psychological makeup manifested in a common culture.”16 Typically, he also informs us that if a single one of these factors is missing, no nation exists. The trouble with definitions of this sort is that they give a false aura of scientific objectivity, which collapses as soon as you start to think of all the nations it would exclude—the United States of America, for one. And although Stalin dismissed the demand for cultural autonomy associated with Austro-Marxism, his definition actually draws heavily on that of Bauer, by retaining the catchall categories of “community” and “psychological makeup.”17

With the triumph of Stalinism in the late 1920s, serious discussion of nationalism virtually ceased. The main source of discussion about the nation therefore passed to non-Marxist political and social scientists, including many who were to be the founding fathers of the academic discipline of International Relations, an orientation that suggests their interests lay in the “state” side of the “nation-state” couplet.18 Yet although they tended to see nationalism as a movement only emerging from the late eighteenth century, they also accepted that nations—at least the “old historic nations” such as Spain, England and France—long preexisted this period.

The Andersonian moment: Political and theoretical contexts

From the 1960s a “modernist” current emerged within the study of nationalism that took a much more foreshortened view of its history. Emphases varied. Of the initial “modernist texts,” Kedourie’s Nationalism (1960) privileged the Enlightenment and Gellner’s essay “Nationalism” (1964) the Industrial Revolution as the sources of nationhood. But all “modernists,” as the name suggests, saw both nations and nationalism as relatively recent, “modern,” creations.19 The intellectual dominance of modernism only held sway for a relatively brief period, roughly from the late 1970s to the early 1990s, and of the writers identified with it, only Tom Nairn, Eric Hobsbawm and Nigel Harris saw themselves as Marxists at the time of writing, as did Anderson.

Anderson is a specialist in East Asian politics. He went to Indonesia in 1962 to study that country’s experience during the Second World War, when Japanese occupation supplanted the Dutch colonial presence. As he later commented, his research suggested that there were parallels between the period of resistance to Japan and the returning Dutch, and the 1960s, when the “Guided Democracy” of President Sukarno stood on the edge of a real social radicalization, only to be brought to a bloody end by the coup of 1965.20 The book that resulted from these researches, Java in a Time of Revolution (1972), dealt only in passing with the question of nationalism, but what it does say is interesting in the light of his later preoccupations. Here, Anderson describes the second congress of the youth wing of the Indonesian National Party in 1928:

At that same congress the participating youth took the historic oath of commitment to one people, the Indonesian people, one nation, the Indonesian nation, and one language, the Indonesian language. In a real sense the development of nationalism among the educated youths was less the result of their readings in European history or their encounters with racist colonial authorities than of their experience in schools whose raison d’être derived directly from the centralised structure of the Netherlands Indies in the twentieth century. Nationalism was the only explanation that could be given for each student having made the journey from whatever home he had left to enter one classroom with the others. Only nationalism made sense of the new life on which they were collectively embarked. But insofar as this nationalism was a response to experience, it was necessarily limited to those who shared that experience. The politically-minded youth of the 1930s were profoundly isolated from the rest of their contemporaries. It was not until the Japanese period that nationalism spread deeply into small-town and rural Java; and it did so then because of the new experiences encountered there, to which it gave coherent meaning.

In describing the later role of the Barisan Pelopor (Vanguard Corps) during the resistance, Anderson notes: “It was a new nationalism, very different from that of their elders, precisely because it grew out of an unprecedented experience.”21 Many of the themes rehearsed here—the initial growth of nationalism as emerging from the collective experience of an elite group, the sense of nationalism as a means of understanding the world rather than a narrow set of political demands—were to re-emerge in more fully developed form in Imagined Communities. Anderson was expelled from Indonesia for displeasing the Suharto regime shortly after the book appeared. “Exile,” he later wrote, “had the advantage of pushing my inquiries back into the nineteenth century, and from everyday politics to the transformations of consciousness that made presently existing Indonesia thinkable.” There were, however, other factors that led him to write Imagined Communities.

In 1978 and 1979 wars had taken place between Vietnam, Cambodia, and China, but “none of the belligerents had made more than the most perfunctory attempts to justify the bloodshed in terms of a recognizable Marxist theoretical perspective.” This said something about their character: “Since World War Two every successful revolution has defined itself in national terms . . . and, in so doing, has grounded itself firmly in a territorial and social space inherited from the revolutionary past.”22 But the idea that socialism, or even the transition to socialism, should perpetuate nation-state and nationalism was contrary to all previous Marxist positions. What implications did this have for the Marxist theory of nationalism? There was already a perception that Marxism lacked an adequate theory of nationalism. In 1976 Tom Nairn had claimed that nationalism “represents Marxism’s greatest historical failure.”23 As Anderson points out, that implies “the regrettable outcome of a long, self-conscious search for historical clarity.” His response was that the problem was rather that this search had never taken place; consequently, “nationalism has proved an uncomfortable anomaly for Marxist theory and, precisely for that reason, has been largely elided, rather than confronted.”24 Anderson now reveals that he was more sympathetic to Nairn’s position than this suggests and claims that he intended Imagined Communities to offer critical support, but also to extend Nairn’s critique from Marxism to all other political traditions, which he saw as similarly lacking.25 What then was his alternative?

Anderson’s argument

Anderson starts by arguing that nationalism is “a radically changed form of consciousness.”26 To define it, he starts with the reason why the nation has to be imagined: “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 community.” Anderson is insistent that “imagined” does not mean “false,” because all communities beyond the original gatherer-hunter groups have to conduct a similar act of imagining: “Communities are to be distinguished, not by their falsity/genuineness, but by the style in which they are imagined.” Anderson argues that there are three aspects to what is being imagined: limitation, because no nation can encompass the entire world and the boundaries of each are set by other nations; sovereignty, because nations came into existence at the time when the legitimacy once conferred by absolutist divine right was being replaced by that of the state; and community, because the horizontal solidarities of the nation were stronger than vertical oppositions, even those of class.27

Anderson identifies “the end of the 18th century” as the period that saw “the spontaneous distillation of a complex ‘crossing’ of discrete historical forces,” and once distilled it was no longer necessary for each potential new nation to have undergone the same experiences. They could be “transplanted, with varying degrees of self-consciousness, to a great variety of social terrains.”28 But the origin of these forces goes much further back in time. Anderson argues that from the late medieval period onward there was the collapse of three key conceptions of the world: the idea that belief systems expressed in particular script languages like those of Christianity and Islam (using respectively Latin and classical Arabic) offered privileged access to truth; the belief that society was naturally organized around and under monarchs who were persons apart from other human beings and who ruled by some form of divine dispensation; and an understanding of the past and present in terms of some creation myth. Such notions rooted human lives firmly in the very nature of things, giving certain meanings to the everyday fatalities of existence (above all death, loss, and servitude) and offering, in various ways, redemption from them. All these conceptions were subverted by economic change, discoveries social and scientific, and the development of increasingly rapid communications: “No surprise then that the search was on, so to speak, for a new way of linking fraternity, power and time together.”29 He is not of course making the absurd suggestion that religion had to decline in importance before nationalism could emerge, contrary to the claims of several of his critics.30 Given the importance he ascribes to the Reformation, this is clearly a misreading of his argument. What he is suggesting instead is that the nature of what people expected from religion changed, at least in the West.

For Anderson, the solution was provided by the emergence of “print capitalism.”31 This created the possibility of a vast market beyond the tiny minority who could understand Latin. Print-languages “created unified fields of exchange and communication below Latin and above the spoken vernaculars.” They “gave a new fixity to language, which in the long run helped to build that image of antiquity so central to the subjective idea of the nation.” And they created “languages of power,” with certain dialects playing a dominant part in communication through printing. These were “largely unselfconscious processes resulting from the explosive interaction between capitalism, technology and human linguistic diversity.”32

The remainder of the book sets out how national consciousness spread and was transmuted into nationalism. He argues there were three main kinds of nationalism, arising in successive waves: “creole” nationalism associated with the revolt of the American colonies (“creole” in its Spanish use means a Latin American of European ancestry); “language” nationalism associated with Western Europe; and “official” nationalism associated with Central and Eastern Europe, and with the Asian and African anticolonial movements. In what is perhaps his boldest innovation, Anderson argues that the “pioneers” of nationalism were the first of these, the colonial states of the Americas. He ascribes the rise of nationalism to the attempt by Madrid to impose greater control, the influence of Enlightenment ideas, and the way in which the South American continent had been divided into particular, territorially delimited administrative units: “In this respect they foreshadowed the new states of Africa and parts of Asia in the mid-20th century . . . The original shaping of the American administrative units was to an extent arbitrary and fortuitous, marking the spatial limits of particular military conquests. But over time they developed a firmer reality under the influence of geographic, political and economic factors.”33

Two necessary internal processes translated the brute fact of territoriality into national consciousness, according to Anderson. There was the self-identification of the descendents of settlers with the colonial territory, in distinction from their European-born equivalents. And there was the emergence of a particular manifestation of print capitalism: the newspaper that “brought together, on the same page, this marriage with that ship, this prince with that bishop,” creating “quite naturally, and even apolitically . . . an imagined community among a specific assemblage of fellow-readers, to whom these ships, brides, bishops and princes belonged. In time, of course, it was only to be expected that political elements would enter in.”34

Although vernacular language was critical to the original formation of national consciousness, once nationalism became available as a model, it was no longer necessary for new nations to have this as their basis.35 The emergence of nationalism was originally associated with the popular masses, but it became available for use for “conservative, not to say reactionary” ends by the state bureaucracies of societies that had not experienced successful popular movements.36 Anderson argues that the anti-imperialist nationalism, which began to build new states after 1945, drew on both of these aspects. “That is why so often in the ‘nation-building’ policies of the new states one sees both a genuine, popular nationalist enthusiasm and a systematic, even Machiavellian, instilling of nationalist ideology through the mass media, the educational system, administrative regulations and so forth.”37

A brief exposition can only hint at the subtlety, complexity, and sophistication of Anderson’s arguments. More than any previous writer, Anderson established that the phenomenon of nationalism was constructed and historical, not natural and eternal. Part of the charm of the book lies in the sheer range and novelty of the examples that Anderson musters to illustrate his argument, many of them drawn from areas such as Burma, Thailand, and Indonesia, which do not normally feature in discussions of nationalism. Reading it again, I was struck by how strange a book it is. The strangeness is partly a matter of style. Anderson refers in several places to Walter Benjamin’s opposition of “Messianic time” and “empty homogenous time,” but the book also reproduces many aspects of the German writer’s allusive and disconnected manner.38 Benjamin was primarily a philosopher and cultural critic, disciplines where an opaque literary technique is not necessarily a disadvantage; in Benjamin’s case, the elliptical nature of his writings consciously invokes the fragmentation of the modern consumerism whose birth he recounted in, among other works, The Arcades Project.39

And yet Imagined Communities is one of those books that are of great individual value, but that have ultimately exerted a negative influence on socialist thought. As I suggested earlier, this is due to problems within the book itself, and not just to misreadings. Given that I strongly disagree with several aspects of Anderson’s account of nationalism (most in relation to what he omits), it is more than usually necessary to state in advance which criticisms of his work or inferences from it I think are wrong or illegitimate.

Postmodernist appropriation and primordialist approbation

Imagined Communities made an immediate impact on publication, but perhaps not in the way that Anderson had hoped. Although other Marxists did find his work useful, it actually provided far greater support for emerging ideologies fixated on questions of identity, above all postmodernism. One of Anderson’s more insightful critics, Anthony Smith, noted that, while Anderson’s project is not itself a postmodernist reading, “it is the idea of the nation as discourse to be interrogated and deconstructed, that has proved most influential.”40 Anderson does not, of course, suggest that the nation is simply a discourse, but many of his critics have found it convenient to ascribe that view to him. The Scottish writer Murray Pittock, for example, writes, “The weakness of Anderson’s notion of the ‘imagined community’ is that it implies that one can imagine at will, and choose an identity as the postmodern consumer chooses a lifestyle product.”41 Again, this position would be nonsense, but Anderson nowhere argues it; on the contrary, his argument is in the adoption of identities that are only available because of a determinate (although certainly not determined) historical process.

What both Smith and Pittock oppose in Anderson’s work is precisely his constructivism, his modernism. But some postcolonial thinkers regard Anderson as too materialist. Partha Chatterjee, for example, writes that “the theoretical tendency represented by Anderson certainly attempts to treat the phenomenon as part of the universal history of the modern world.” He opposes this in relation to Africa and Asia:

If nationalisms in the rest of the world have to choose their imagined community from certain “modular” forms already made available to them by Europe and the Americas, what do they have left to imagine? History, it would seem, has decreed that we in the postcolonial world shall only be perpetual consumers of modernity. Europe and the Americas, the only true subjects of history, have thought out on our behalf not only the script of colonial enlightenment and exploitation, but also that of our anti-colonial resistance and postcolonial misery. Even our imaginations must remain forever colonised.

Chatterjee argues that “anti-colonial nationalism creates its own domain of sovereignty within colonial society well before it begins its political battle with the imperial power.” This domain is the spiritual one, “an ‘inner’ domain bearing the ‘essential’ marks of cultural identity.”42 Anderson did not respond directly to Chatterjee, but in a later article for New Left Review he rightly argued that the distinction between Western and Eastern nationalism was relatively unimportant: “The oldest nationalisms in Asia—here I am thinking of India, the Philippines, and Japan—are older than many of those in Europe and Europe Overseas—Corsica, Scotland, New Zealand, Estonia, Australia, Euskadi, and so forth.” Anderson has no difficulty in demonstrating that creole, linguistic, and official types of nationalism are as prevalent in the East as in the West. And, as he also notes, “what people have considered to be East and West has varied substantially over time.”43

Anderson is therefore not vulnerable to criticism on the grounds Chatterjee raises. But he is prevented from making a more vigorous response because of a flaw in his own theory, namely the contingent role it allocates to capitalist development. I will return to the point below, but it is possible here to make a direct answer to Chatterjee. If India had achieved rounded capitalist development early enough, such that it was able to resist British imperialist aggression (an outcome far more conceivable there than in, for example, China), something like nationalism as it is understood in “Western” terms would have emerged because it is a necessary superstructural component of capitalist societies. What Chatterjee is proposing is an ahistorical mysticism that, in an interesting example of what Hegel used to call the “interpenetration of opposites,” actually reproduces in “postcolonial” guise many of the criticisms raised by more conventional critics.

Smith is one of the main proponents of the “perennialist” view of nations that sees them as rooted in much older ethnic identities. He considers it a problem that, for Anderson, “the nation possesses no reality independent of its images and representations. But such a perspective undermines the sociological reality of the nation, the bonds of allegiance and belonging which so many people feel, and obscures both the institutional, political and territorial constitution of nations, and of the powerful and popular cultural resources and traditions that underpin so many nations and endow them with a sense of tangible identity.”44 Similarly, Pittock wants to defend a conception of Scottish nationalism stretching back to early medieval times: “The idea that identities are mainly chosen, invented or dreamt up is, like the teleological and sometimes Marxian history whence it derives, a judgement on the past in terms of a present agenda: ‘be whatever you want to be’ may be a sentiment understood by our posterity, but it was inaccessible to our ancestors.”45 Unfortunately, nationalist identities were also inaccessible to our ancestors, whatever fantasies may be entertained about the Declaration of Arbroath, or whatever. In other words, Anderson is being attacked here precisely where his work is strong, not where it is weak.

Nevertheless, as these criticisms demonstrate, the dominance of the modernism to which Anderson belonged was relatively brief. Politically, the latest resurgence of nationalism, and theoretically, the idiocies of postmodern identity politics, have given impetus to a revival of arguments that hold that nations are much older than modernists have claimed. As Pittock notes, “the idea that nations and nationalism cannot predate the French Revolution” is increasingly on the defensive.46 The final collapse into primordial essentialism was signaled by two works by Emmanuel Todd, L’illusion Economique (1998) and La Diversité du Monde (1999). Todd’s key thesis has been approvingly summarized by Tom Nairn to claim that “nationalism is constitutive of man’s social nature.”47 Naturally Nairn has no time for the constructivist interpretation of Anderson: “Peoples have not ‘imagined’ such communities by chance, or out of irrational impulses from the soul. Identities are not aesthetic choices, but ways of existing.” He therefore endorses what he takes to be a change of heart by Anderson in a later essay called “The Goodness of Nations.”48

Reimagining the history of the nation

Imagined Communities can still play a role in intellectually challenging both postmodernism and primordialism, but only if its themes are integrated into a more consistently materialist framework. The book consists of a series of impressionistic studies on particular aspects of nationalism. But the connections between them are often difficult to establish. What is missing is any central dynamic linking them together, except for the concept of print capitalism. Yet the universality of contemporary nationalism suggests that it was originally produced and subsequently reproduced by a set of conditions wider and more fundamental than this. Anderson’s argument about the coincidence of existential doubt and technological advance in print seems unconvincing as an explanation for something as all-pervasive as nationalism. A more convincing explanation might be the more general development of capitalism and, given the way in which antimodernists have attempted to deny any necessary relationship between nationalism and capitalism, we might have expected Anderson to argue the contrary.49 But, like Otto Bauer, Anderson sees a purely contingent relationship between this and the rise of nationalism: “What made the new communities imaginable was a half-fortuitous, but explosive, interaction between a system of production and productive relations (capitalism), a technology of communications (print), and the fatality of human linguistic diversity.”50 He made the same point, even more strongly, in a later essay: “The two most significant factors generating nationalism and ethnicity are both linked closely to capitalism. They can be described summarily as mass communications and mass migrations.”51 The connection between national consciousness and capitalism is, however, far more all-embracing than this suggests.52 In fact, national consciousness took as many centuries to become the dominant form of consciousness as the capitalist mode of production did to become the dominant mode of production, and it did so as a consequence of that. Four main elements combined, reflecting to a greater or lesser extent the impact of capitalism on feudal society.

The first element was the formation of externally demarcated and internally connected areas of economic activity. Europe had emerged from the first crisis of feudalism by the later fifteenth century as a system of states that was still dominated by the feudal mode of production. It was a system, however, increasingly adapted to elements of capitalism. In this context, the importance of capitalist development was less in the domain of production than that of circulation, for it was in the creation of trade networks that merchant capital began to link up dispersed rural communities both with each other and with the urban centers to form an extensive home market.

Linked directly to this element was a second, the adoption of a common language by the communities that were being connected to each other at the economic level. The need to communicate for the purposes of market exchange began to break down the distinctiveness of local dialects, forging a language common, or at least comprehensible, to all. Language in this way began to set the boundaries of the economic networks referred to above, boundaries that did not necessarily coincide with those of medieval kingdoms. Such economic and linguistic unification was far easier in a small centralized kingdom such as England than in a territory such as the German Empire. Indeed, establishment of state frontiers often purely determined the boundary between a dialect of a particular language and another language. And of course Anderson is right that the formation of standard forms of language was immeasurably aided by the invention of printing and the possibilities it presented for the codification of language in mass-produced works. These would not have been produced unless an audience of the literate already existed that understood their contents, but their effect was to extend the size of that audience, since printers could not produce works in every local dialect, only in the one that had emerged as the standard form, or in those that were in competition to do so. The increasing standardization of language then fed back into its original economic formation, as the merchants whose trading networks had originally defined the territorial reach of linguistic comprehensibility increasingly identified themselves with that territory, to the exclusion of rivals who spoke a different language. The rise of the vernacular was accompanied by the decline of Latin as a lingua franca, a process virtually complete by the mid-sixteenth century and expressed in the new profession of interpreter, now necessary to make vernacular diplomatic exchanges mutually comprehensible.

The third element was the character of the new absolutist states. Absolutism was the form taken by the feudal state during the economic transition from feudalism to capitalism. Yet the absolutist states did not arise automatically. The replacement of the estates monarchy of the earlier feudal period by a more centralized apparatus was the political response of the feudal ruling class to the social and economic pressures—different in degree and combination throughout Europe—set in train by the first crisis of the feudal system and the greater significance of capitalist production in the economies that emerged from it. The local jurisdictions that characterized the classic epoch of military feudalism began to give way to greater concentration of state power, notably through the introduction of standing armies and, partly in order to pay for them, regular centralized taxation.

Death and taxes both involve bureaucracies that require a version of the local language comprehensible across the state territory, thus strengthening the “linguistic” element. They also had two unintended effects. The introduction of regular taxation and the adoption of mercantilist policies reinforced the economic unity that had begun to emerge spontaneously from the activities of merchant capitalists. And the military rivalry that characterized the new system necessitated mobilizing the active support of the bourgeois minority as a source of financial backing and administrative expertise. Despite these innovations it is nevertheless important not to mistake the role of absolutism in the birth of nationhood, which was that of a midwife, not that of a mother. The issue is often elided by reference to the influence of “the modern state” in the creation of nations, but this is to dissolve the difference between the absolutist state and its genuinely modern bourgeois successor. The arrival of nationhood coincided not with the establishment of the absolutist states but with their overthrow.

The fourth and final element is local manifestations of a global religious belief. The ideology of absolutism involved stressing the deeds of religious figures such as saints, who were associated with the territory of the realm, but it was the Reformation that made religion more than an ideologically pious enhancement to the image of the ruling dynasty. Wherever Protestantism became the dominant religion within a given territory after 1517 it contributed to the formation of national consciousness by allowing communities of belief to define themselves against the interterritorial institutions of the Roman Catholic Church and the Holy Roman Empire. In part this was through the availability of the Bible in the vernacular, but this in turn depended on the existence of linguistic frameworks in which market transactions and state administration could be carried out. Protestantism acted as a stimulus to national consciousness only to the extent that the development of capitalism had provided it with the framework to do so. Naturally the process went furthest in England, but even there it was not until after the death of Elizabeth in 1603 that Protestantism came to be separated from regnal solidarity with the monarch. It took longer for Catholicism to play the same role.

There is therefore a problem with Anderson’s focus on “creole” nationalism as the major formative experience of nationalism. Apart from anything else, he contradicts himself by describing it as drawn from an earlier model: “In effect, by the second decade of the 19th century, if not earlier, a ‘model’ of ‘the’ independent national state was available for pirating.” He describes this as “a complex composite of French and American elements.”53 But incredibly, this is one of the first occasions that France is mentioned. To ignore the influence of the French Revolution in establishing the “model” seems particularly perverse. The problem here is that even France and the United States are not the first nations. The United Netherlands and England have a stronger claim to priority. To argue that nations only appeared at some stage in the later eighteenth century would be as absurd as arguing that capitalism only appeared at the same period. While Anderson is right to draw attention to cumulative movements, he misses something else, which is the explosive effect of the revolutionary turning points that punctuate capitalist development, and their impact in coalescing hitherto inchoate ideological elements into a national identity. His account is, so to speak, all process and no events.

The success of groups with an emergent national consciousness in the Netherlands and England in elevating this new form of consciousness into political movements led others (first in North America, Ireland, and France, then generally) to aspire to national status, even if their level of social development had not previously allowed national consciousness to arise. The bourgeois revolutions effected the final transformation of the term “nation” to one that stood for “the people” as a community—although one of the most divisive issues within all bourgeois revolutionary movements was precisely how “the people” should be defined. The struggle against absolutism required the mobilization of at least a large minority of “the people” to achieve the expulsion or destruction of the royal dynasty. This could only be done by providing some form of identity that could embrace the often very different forms of opposition to the crown, regardless of whether the ruler in question was foreign (as in the case of Spanish Habsburg dynasty in the Netherlands) or native (as in the case of the Stuart dynasty in England). Nationalism provided this identity.

National consciousness could not flourish, or even take root, unless the conditions for capitalist development were present, and for it to be consolidated across Europe, even if only among the bourgeoisie, there had to be at least one case where it made the transition to nationalism and then became embodied in a nation-state. Only when there were concrete examples of nationhood could different groups know what they were conscious of, regardless of whether they then went on to develop nationalisms of their own or not.

The capitalist nation-state became a permanent feature of the international state system only toward the end of the hundred years between the end of the English Revolution in 1688 and the beginning of the French Revolution in 1789. Thereafter new nations could be manufactured regardless of whether the original elements were present or not—although an economic infrastructure and common language would, of necessity, have to be introduced at some point for a sense of national consciousness to be consolidated. The ideological dominance of nationalism over the population depended, however, on when a particular revolution occurred in the overall cycle of bourgeois revolutions. In the two states where bourgeois revolutions were successfully completed before or during 1688, the Dutch and the English, the existence of national consciousness was directly proportional to the extent that the postrevolutionary state developed a centralized apparatus, rather than a federal or confederal structure. In this respect English nationalism was as far in advance of its Dutch predecessors as it was of its American successor, which similarly remained an alliance of semiautonomous states down to 1865.

After 1848 all ruling classes intent on creating states on the British or French models were forced to embrace nationalism, not because they were personally capitalists, but because all of them—Prussian Junkers, Japanese Samurai, Italian monarchists, and, eventually, Stalinist bureaucrats—were engaged in building industrial societies dominated by the capitalist mode of production. The example of Italy is typical of how ruling classes were faced with the need to diffuse consciousness of being a nation down from elite level into the mass of the population, a large and growing proportion of whom were not the bourgeoisie and petty bourgeoisie who had originally formed the nation, but workers. The difficulties involved should not be underestimated: as late as the 1860s as many as a quarter of the inhabitants of the French state did not speak French.

Class is the great absent theme in Imagined Communities, yet no Marxist account can deal with the subject without exploring the role nationalism plays in class relations. National consciousness begins to emerge in the social classes below the rulers of the new nation-states, partly as the result of deliberate indoctrination, but far more so as the by now inevitable pattern of life experience within societies shaped by the nation-state form. Among the working class the existence of reformist class consciousness provides the context within which national consciousness and nationalism develop. Reformist class consciousness was originally a historical product of the social conditions produced by the transition to capitalism or, more precisely, by the process of capitalist industrialization, first in Britain and subsequently elsewhere.

Once the initial shock of industrialization passed, workers came to accept that capitalism was not a passing aberration but a new form of society that might have many years of vitality ahead of it. The apparent permanence of the system forced accommodation and adaptation, however grudgingly, from the new exploited class, whose horizons were anyway limited by the “dull compulsion” to work, raise families, and recover from the savage exertions demanded by the factory system. Although these conditions provoked resistance, the fact that the new system generated its own defensive illusions made the possibility of a generalized revolutionary class consciousness emerging out of these resistance struggles less likely. Under early capitalism exploitation was accompanied by the economic discipline instilled by fear of the poverty that would result from being sacked. The actual process of exploitation, the fact that the worker produced more than that for which she or he was rewarded, was hidden from view. As a result, although workers were usually hostile to their own particular boss, this did not necessarily generalize into opposition to the system as a whole. Although trade unions grew out of worker resistance, the goal of these new organizations, whatever rhetoric was employed about the (invariably distant) overturning of the system, was improving the condition of the working class within the system itself. The resulting contradictory form of consciousness finds its most basic expression in an acceptance by workers of the wages system accompanied by a rejection of the particular level of wages they are being offered, but it extends to all aspects of social life.

What then is the relationship of national consciousness to this reformist consciousness? National consciousness does not compete with revolutionary class consciousness directly for the allegiance of workers, but as a key element in reformist class consciousness. Indeed, one might say that workers remain nationalist to the extent that they remain reformist. And from the point of view of the capitalist class in individual nations it is absolutely necessary that they do so, or the danger is always that workers will identify, not with the “national” interest of the state in which they happen to be situated, but with that of the class to which they are condemned to belong, regardless of the accident of geographical location. Nationalism should not therefore be seen as something that only “happens” during separatist movements on the one hand, or during Fascist and imperialist manifestations on the other. The capitalist system generates nationalism as a necessary everyday condition of its continued existence.

Mass nationalism was therefore initially a product of industrialization, but not simply because it is functional for the ruling class in industrial capitalism. Industrialization and urbanization together produced the changes in human consciousness that made nationalism possible (for the subordinate classes), as well as creating societies that made nationalism necessary (for the dominant class). They developed new structural capacities, new modes of experience, and new psychological needs in the people who had to work in the factories and live in the cities. Nationalism responds to the need for some collective sense of belonging with which to overcome the effects of alienation, to the need for psychic compensation for the injuries sustained at the hands of capitalist society, in the absence of revolutionary class consciousness, but in conjunction with reformist class consciousness.

The ideological role played by the ruling class in reinforcing nationalism is therefore only possible because nationalism already provides one possible means of meeting the psychic needs created by capitalism. Once a capitalist nation-state has been established, those who control the apparatus always seek to consolidate the hold of nationalism among the people who inhabit its territory. States need conscripts for their armies, citizens to pay taxes, workers to accept that they have more in common with those who exploit them at home than they do with their fellow-exploited abroad. This made it imperative that loyalty to a state be secured, and the nation was the means. Since the eighteenth century British workers have often been asked to accept rises in interest rates, cuts in wages and services, or participation in imperialist wars, but never for the benefit of British capitalism, always for the benefit of the British nation, for “the national interest.”

It is not only the state that makes such appeals. The organizations of the working class themselves reinforce reformist class consciousness within a national context. At the most elementary level this is because such organizations are unwilling to challenge the nationalism within which political discourse is conducted, for fear of being labeled unpatriotic. More importantly, however, it is because they seek either to influence or to determine policy within the confines of the existing nation-state. Typically, therefore, nationalism is invested with the contradictory character of the reformist worldview.

Conclusion

If nationalism is as intertwined with capitalism as I have suggested, then nationalism today can only ever be progressive in certain limited circumstances, most obviously in relation to movements against national oppression. Marxists cannot be nationalists, nor can they even support nationalisms as such, although they can support particular national demands or movements. This is not Anderson’s conclusion, as he hinted in Imagined Communities itself: “In an age when it is so common for progressive, cosmopolitan intellectuals (particularly in Europe?) to insist on the near-pathological character of nationalism, its roots in fear and hatred of the other, it is useful to remind ourselves that nations inspire love, often profoundly self-sacrificing love.”54 He made the point more clearly in a lecture given in the Indonesian capital of Jakarta shortly after the overthrow of the Suharto regime. Anderson told his audience that he had “half-jokingly” raised the slogan “Long live Shame!”

No one can be a true nationalist who is incapable of feeling “ashamed” if her state or government commits crimes, including those against her fellow citizens . . . During the Vietnam War, a good part of the popular opposition came from just this good sense of shame among the American citizenry that “their government” was responsible for the violent deaths of three million people in Indochina, including uncounted numbers of women and children . . . So they went to work in protest, not merely as advocates of universal human rights, but as Americans who loved the common American project.55

The distinction between true and false nationalists is dangerous in the extreme and opens up the possibility of collapsing into old Stalinist notions about “reclaiming our history” from the right, “progressive patriotism,” and so on.

In the concluding chapter to the original edition, Anderson expressed the forlorn hope that the experience of what he called “inter-socialist wars” would not be repeated: “But nothing can be usefully done to limit or prevent such wars unless we abandon such fictions as ‘Marxists as such are not nationalists’ or ‘nationalism is the pathology of modern development,’ and, instead, do our slow best to learn the real, and imagined, experience of the past.”56 Like Nairn before him, Anderson then goes on to invoke Walter Benjamin’s image of the Angel of History, propelled into the future by a storm from Paradise, the storm called Progress, while the debris of history gathers beneath him.57 It is high time we stopped the legacy of this great, if eccentric, Marxist thinker being misappropriated in ways that are quite alien to his thought. Benjamin understood the tragedy of history as few other Marxists have done, but he did not see the future of humanity as necessarily being an endless prolongation of that tragedy. On the contrary, he considered it necessary for there to be a “messianic interruption” of history to bring merely capitalist progress to a halt: “A genuinely messianic face must be restored to the concept of a classless society and, to be sure, in the interest of furthering the revolutionary politics of the proletariat itself.”58 The messianic vision to which Benjamin alludes involves something more than looking for the best in existing nations or seeking to create new ones on a capitalist basis. Anderson’s important work, illuminating in so many ways, is ultimately a failure because he remains trapped within the ideological presuppositions of its subject. He has shown us that national consciousness and the nation-state are forms whose beginnings can be found in past history. But because he misunderstands the forces that brought them into being, he fails to recognize that future history may also see them brought to an end.