From Progress to Catastrophe

Perry Anderson on the historical novel

Within the huge multiverse of prose fiction the historical novel has, almost by definition, been the most consistently political. It is no surprise that it should have occasioned what is still probably the best-known of all works of Marxist literary theory, Lukács’s The Historical Novel, written in Russian exile in the 1930s. Any reflection on the strange career of this form has to begin there, however far it may then wander from him. Built around the work of Walter Scott, Lukács’s theory makes five principal claims. The classical form of the historical novel is an epic depicting a transformation of popular life through a set of representative human types whose lives are reshaped by sweeping social forces. Famous historical figures will feature among the dramatis personae, but their roles in the tale will be oblique or marginal. Narratives will centre instead on middling characters, of no great distinction, whose function is to offer an individual focus for the dramatic collision of opposing extremes between whom they stand, or more often waver. What Scott’s novels then stage is a tragic contest between declining and ascending forms of social life, in a vision of the past that honours the losers but upholds the historical necessity of the winners. The classic historical novel, inaugurated by Waverley, is an affirmation of human progress, in and through the conflicts that divide societies and the individuals within them.

It follows from Lukács’s conception that the historical novel is not a specific or delimited genre or subgenre of the novel tout court. Rather, it is simply a path-breaker or precursor of the great realistic novel of the 19th century. A generation later, Balzac – for example – essentially adapted Scott’s techniques and vision of the world to the present instead of the past, treating the France of the Restoration or the July Monarchy in much the same way that Scott had represented mid-18th-century Scotland or 12th-century England. Balzac’s great successor, for Lukács, was the towering figure of Tolstoy, whose War and Peace represents a peak simultaneously of the historical and of the realist novel in the 19th century. In societies more advanced than Russia, on the other hand, the development of capitalism had by this time pitted a revolutionary working class against a bourgeoisie that no longer believed it bore the future within it, and was intent on crushing any sign of an alternative to its rule. In this quite different – but after 1848 much more typical – situation, the connections of the past with the present were cut in European fiction, and the historical novel gradually became a dead antiquarian genre, specialising in more or less decadent representations of a remote past with no living connection to contemporary existence, but functioning rather as a rejection and escape from them. Such was, archetypically, the fantasy of ancient Carthage constructed by Flaubert in Salammbô.

Fredric Jameson, while remaining faithful to Lukács’s overall vision, has offered a different periodisation within it. He suggests that, rather than seeing Scott as the founder of the classically realist historical novel, we should view him as the practitioner of a costume drama whose narrative form stages a binary opposition between good and evil. A naive ethical antithesis of that sort is the mark of melodrama, and it is no accident – Jameson suggests – that its characteristic artistic expressions should be operatic, rather than novelistic. The truth of Scott is to be found in Rossini or Donizetti, who borrowed from him, rather than in Manzoni, let alone Balzac.

If we wish to see a truer exemplar of Lukács’s prescriptions, Jameson argues, we must turn to War and Peace, a historical novel whose triumph is to transcend the costume-drama oppositions of hero and villain, in its remarkable portraits of a hyper-active yet futile Napoleon, brittle symbol of the French, and an apparently torpid yet supremely sagacious Kutuzov, authentic representative of the slow rhythms of the Russian people and the peasant masses composing its overwhelming majority. Not only this. In War and Peace we find a realism so advanced in its figurations of psyche and sexuality that we seem to be in the presence of a modernism avant la lettre. Yet there is a paradox here, Jameson remarks, in that modernism proper, because of its commitment to the primacy of immediate perception, appears to have been constitutively incapable of generating the totalising retrospect that defines a true historical novel.

This is a brilliant rewriting of Lukács, the seduction of whose starting point – recasting Scott as a librettist – is difficult to resist. Certainly the ethical binary of good and evil imposes a logic of melodrama on much of Scott’s work, but it might be said that Jameson has flinched from the conclusion to be drawn from his own argument. For the fact is surely that Tolstoy – far from transcending this melodramatic structure – reproduces it still more extravagantly in the public narrative of War and Peace, where the opposition between Napoleon and Kutuzov takes the binary of villainy and virtue, the hateful versus the heroic, to caricatural extremes that Scott rarely ventured. Tolstoy’s portrait of Napoleon, Jameson concedes, is perhaps rather petty, unduly diminishing the emperor. But he implies that Tolstoy can thereby be acquitted of demonising him. An answer to this, however, can be found in Sartre’s notable deconstruction of the concept of evil in Saint Genet, where he shows that the pure notion of evil is always caught in the same aporia – what he calls a tourniquet. Either supreme iniquity requires a hideous intelligence and daring in the performance of great crimes, but if so it thereby borrows qualities from the good and ceases to be absolutely evil. Or it is a merely deadened violence, devoid of any reflective, let alone heroic dimension, but in that case it becomes a pathology without responsibility, whose moral insignificance – ‘banality’, Arendt would call it – fails in the opposite way to represent true evil. Tolstoy’s belittling of Napoleon is simply the second variant of this antinomy of villainy, in itself no better than the first.

That this is so can be seen vice versa in the figure of Kutuzov. Jameson suggests that he is no conventional hero, but an empty actant whose function is simply to represent the Russian peasant masses. What tells against this view is not just the effusive, sentimental detail Tolstoy lavishes on the serf-owner general – far more a projection of his own ideology than anything to do with the Russian peasantry – but the lengths to which, as Viktor Shklovsky demonstrated, he went to rig the historical record in constructing him as a patriotic icon. In preparing War and Peace, Tolstoy took the trouble to study, however selectively, materials of the period, so was not unaware of salient realities of the time, but deliberately bent the evidence where it was inconvenient for his propagandist purpose. This is something Scott was on the whole reluctant to do. In War and Peace we are closer to the spirit of Alexandre Dumas’s maxim: ‘On peut violer l’histoire à condition de lui faire de beaux enfants.’

The question, of course, is whether Tolstoy’s fictional portrait of Kutuzov qualifies as such a handsome offspring – that is, a persuasive work of art. The evidence that it fails to do so is written out in extenso in the novel itself, whose incoherent philosophical tirades on the nature of history – deplored by virtually all its readers – function as a compulsive make-weight for the flimsiness of the oleograph at one centre of the narrative, the political stage on which the fate of Russia is played out. The personal destinies of its fictional characters are another matter. But to grasp the sense in which War and Peace is a historical novel, classically interconnecting public events and private lives, it needs to be reinserted in the series of which it is a member. This is something Lukács’s account of the form touches on, but then skirts. The historical novel – if we except its one great precursor, Kleist’s Michael Kohlhaas – is a product of romantic nationalism. This is as true of Tolstoy as it is of Scott, Cooper, Manzoni, Galdós, Jókai, Sienkiewicz or so many others.

The original matrix of this nationalism was the European reaction against Napoleonic expansion. Although this had popular roots, above all in the Tyrol, Spain and Germany, it was everywhere also driven by the need of the continent’s different ancien regimes to mobilise local enthusiasm for the defence of crown and altar. Viewed as a whole, its dominant was always a counter-revolutionary response to the French Revolution. But each particular national situation generated its own distinct forms of imagination and retrospect. Scott was writing in the one bastion of the old order that went unscathed through the Napoleonic Wars, without so much as a single French footprint on British soil. The focus of his nation-building narrative is thus quite distinct from its Continental sequels, and corresponds to the duality of the composite British kingdom itself.

On the one hand, there is the heroic story of the emergence of English national identity, as it first took shape in the struggle of democratic Saxons against aristocratic Normans in the early Middle Ages – Ivanhoe – and then deepened and developed through the early modern Tudor and Stuart periods. Here romantic medievalism was given full rein, in fictions rife with melodramatic contrasts and moralising stereotypes of the kind that rightly draw Jameson’s strictures. In European terms, this was probably the most influential side of Scott’s work. On the other hand, Scott was also the chronicler of the peculiar trajectory of his native Scotland, a quite distinct society, within this larger story. Here a completely different vision was at work, formed not so much by Romantic Schwärmerei, as by the Scottish Enlightenment, whose theories of historical development as a universal succession of stages, passing from hunter-gatherer to pastoral to agricultural to commercial forms of society, Scott absorbed from Adam Ferguson and Dugald Stewart and mapped onto his re-creation of the conflicts between Highlands and Lowlands, clan and capital, in Waverley and its sequels. It was Scott’s capacity to represent the tragic collision between historically distinct times and their characteristic social forms – what Bloch would later call Ungleichzeitigkeit – that commanded Lukács’s special admiration, and lifts this part of Scott’s fiction above the moralisms of the costume drama.

In Tolstoy’s case, we can find a very similar sense of the tragic collision of non-synchronous worlds in his tales of tsarist penetration of the Caucasus, above all in his last writing, Hadji Murad, the masterpiece of his historical fiction. There we are shown, with an impassive, laconic tautness closer to Babel or Hemingway than to any writer of his own time, the utterly contrasting worlds of Russian imperialism and Chechen and Avar clan and religious resistance to it, with their own divisions – each side fully realised, with a magnificent economy of means, in a tale as modern as the carnage in Chechnya today. In War and Peace, however, there is an all but complete absence of any drama of this kind, since the conflict that is nominally staged lacks any truly imagined adversary. Tolstoy, in effect, makes no serious attempt to represent the French invader, but seeks rather to obliterate it by devices of minimisation that stretch from a jejune cartoon of Napoleon to voluble protestations that the Grande Armée’s whole expedition was a meaningless affair. The consequence, inevitably, is something close to a chauvinist tract, in which the enemy remains essentially abstract, not a concrete figuration of two contending historical forces. It is partly for this reason that War and Peace, though undoubtedly (and despite Tolstoy’s own denials) a historical novel, set in a period before the author’s lifetime, is so seldom considered primarily as such today.

But there is another reason too. Setting aside the insubstantiality of the French, even within the Russian society it depicts the novel offers little sense of the passage of historical time: that is, of any disconcerting contrast between the epoch of Alexander I, in which it is set, and of Alexander II, in which it was written. At most, the use of French among aristocrats acts as a generic signifier of difference. Otherwise, we are in a kind of continuous present, a more or less eternal Russia in which the leading characters perform much as if they were contemporaries of the author. It is enough to look at Anna Karenina to see how far we are in the same universe. War and Peace thus offers the curiosity of a historical novel with a very weak sense of history, not because Tolstoy was incapable of one – Hadji Murad shows the opposite – but because at this point he was programmatically committed to treating major historical conflicts as meaningless, in the homemade philosophy set out at such length in the book itself.

What lifts the novel above the level of an idiosyncratic – one might say: peculiarly know-nothing – version of romantic nationalism is not just Tolstoy’s extraordinary gifts of observation, and sense of panoramic construction, but the analytic psychology he learned from Rousseau and (above all) Stendhal, whose role in his development as a writer might be compared with that of Ferguson and Stewart in Scott, as the Enlightenment antidote controlling and redeeming the melodrama of national salvation. The greatness of War and Peace lies here: in the side of Tolstoy, closer to Laclos and Stendhal, that was relentlessly rationalist in his dissection of motive and feeling. Of course, there is no clear-cut line of division between the two registers of his writing. Much that is conventionally ‘romantic’ in a mediocre sense can be found even in his individual characterisations, brilliant as many of these are. Natasha, for example, has many of the traits of a chocolate-box heroine, the ‘poetic, graceful imp’ (inspired, his sister-in-law reported, in part by Mrs Braddon’s ‘sensation novel’ Aurora Floyd), whose fantasy function – at the time and since – as winsome vehicle of Russian charm continues to produce such kitsch as Orlando Figes’s pot-boiling celebration of the national culture, Natasha’s Dance. Pierre’s lumbering progress into her arms is the weakest, because most moralistically predictable, feature of the Bildungsroman that the novel also contains. But such impurities are only a reminder of a larger, more constitutive heterogeneity, not so much of this work as such, but of the genre of the historical novel to which it can still be said to belong.

For if the historical novel began as a nation-building exercise in the backwash of romantic reaction to the French Revolution and Napoleonic expansion, the results varied according to context. The curious emptiness of the political space of War and Peace is also a function of the fact that the Russian Empire was an already constituted great power, which victory in 1812 simply conserved as it was. Behind the novel was not an impulse of emancipation, but of displaced compensation. Smarting from Russian humiliation in the Crimean War, where the troops of Napoleon III had stormed Sebastopol under his eyes, and detesting the liberal intelligentsia that blamed the country’s recent defeat on the backwardness of a society still resting on serfdom, Tolstoy fixed on victory over Napoleon I as a salve for contemporary morale, and a rebuff to the callow reformers of 1856 (‘treasonable hands that would not hesitate to ignite the fires of revolt’), who had called for a prompt emancipation of the serfs that he believed could only ‘end in them slaughtering us’. Central to the purpose of the novel became the counter-projection of a fictive unity of peasantry and aristocracy, repelling the foreign invader despite a Frenchified court and (in reality) all too Germanic high command, to be parodied or ignored. Historically speaking, the sublimation of a retrospective revanchism could only yield what Shklovsky called the canonisation of a legend. War and Peace remains a great novel; unlike most other masterpieces of conservative imagination, however, fissured and weakened by its political intentions, not strengthened by them.

Elsewhere, writers were less ideologically driven. In Germany, Fontane started a historical novel about the campaign of 1812-13, Vor dem Sturm, a year before Tolstoy published his. It even includes the memories of a Prussian officer who had fought at Borodino on the other side, with the Grande Armée. But Fontane’s patriotism was free of any jingoist spirit, respectful of French adversaries and critical of Prussian defects. The Prussian reformers, learning the lessons of defeat by Napoleon at Jena, had started to modernise the state as Alexander I did not. Fontane’s pendant to Tolstoy thus inverts its schema. Its aristocratic heroes raise French-style volunteers to fight the French as they retreat from Russia, only to find that, in a dialectical irony, habits of Prussian discipline still make this militia too rigid to prevail against them. In Spain, Galdós’s Episodios nacionales likewise followed the struggle against Napoleon, but could move to the liberal revolution and absolutist reaction that was intertwined with it in a way that Tolstoy could not. By any measure the greatest monument to the form ever produced – a chain of 46 novels written over 40 years – it was set off by the disappointments of the Spanish Revolution of 1868, whose roots Galdós traced to the war that had reclaimed the nation’s independence. In a panorama unsparing of his country’s defects, without a hint of sentimentality about its nobility or its peasantry, or the frailties of its reformers, a reasoned patriotism and pessimism were conjoined.

In Northern Italy, where French rule was often more appreciated than detested, there was no national reaction against the First Empire, so Manzoni – author of a famous ode to Napoleon – had to situate I Promessi Sposi much further back in time, during Spanish rule over Milan in the 17th century, avoiding any too remote antiquarianism while offering a parable of popular life to stir patriotic feeling against Austrian dominion in the time of the Holy Alliance. The logic of this international pattern can be seen a contrario from the case of France, where for obvious reasons no comparable historical narrative could be constructed. Strictly construed, the nearest equivalent – noted by Lukács – would be the regional drama of Balzac’s Chouans, a Vendéen version of Scott’s Highlanders. But the centrepiece of romantic historical fiction in France was, of course, Notre Dame de Paris, whose medieval phantasmagoria, free-floating sentimentalism and detective story motifs place it completely outside the ranks of the classic historical novel as Lukács defined it. Yet this oddity presaged a more pregnant multiplication of the genre.

In the next generation, France became the leading exporter of the costume drama sans phrases, with the extraordinary career of Alexandre Dumas, though England was not far behind with Harrison Ainsworth and G.P.R. James. It was at this point that the historical novel started to acquire its modern ambiguity. Most literary genres have included a variety of registers, and as the Russian Formalists always emphasised, their vitality has typically depended on interactions between high and low, elite and popular forms, either in a circuit within the genre, or via diagonal connections across them. At the same time, the dominant pole within a genre will usually be fairly clear-cut – Symbolist poetry, let’s say, lying at the elite end of the spectrum, thrillers at the popular end. The peculiarity of the historical novel, however, has been to elude any stable stratification of high and low. Its evolution exhibits rather an oscillating continuum of registers, including – to use for a moment anachronistic terms – not just ‘highbrow’ and ‘lowbrow’ but also importantly ‘middlebrow’ ranges of work. It is the extent of this continuum that arguably sets it apart from other narrative forms.

The reason this may be so appears to lie in the nature of its subject matter. For Lukács, the historical novel was essentially epic in form. It was an extensive representation, in Hegelian terms, of the ‘totality of objects’, as opposed to the more concentrated ‘totality of movement’ proper to drama. But if this is a plausible description of the origins of the form, it cannot account for its diffusion. There, it was not an aspiration to epic totality that would ensure the enormous popularity of fictions about the past, but rather the pre-constituted repertoire of scenes or stories of that history, still overwhelmingly written from the standpoint of battles, conspiracies, intrigues, treacheries, seductions, infamies, heroic deeds and deathless sacrifices – everything that was not prosaic daily life in the 19th century. Here was the road, so to speak, from Jeanie Deans to Milady. The historical novel that conquered European reading publics in the second half of the 19th century would not offend patriotic sentiment, but no longer had a nation-building vocation. The Three Musketeers and its innumerable imitations were entertainment literature.

Side by side with it, persisted ‘high’ forms of the genre. Now, however, the typical development was for leading authors to try their hand at the historical novel, composing one or two such works in a corpus otherwise devoted to realistic representations of contemporary life. Barnaby Rudge and A Tale of Two Cities, Henry Esmond, Romola and Salammbô illustrate this pattern. Lower down, but still above the stratum of Dumas or Ainsworth, figured writers like Stevenson and Bulwer-Lytton. The central fact to grasp, however (the evidence for this is graphically laid out in Franco Moretti’s Atlas of the European Novel), is that the historical novel as a genre predominated massively over all other forms of narrative down to the Edwardian era. It combined enormous market success with continuing aesthetic prestige. In the last season of the Belle Epoque, Anatole France was publishing Les Dieux ont soif, Ford Madox Ford his Fifth Queen; even Conrad would end his career with a couple of historical fictions, set once more in Napoleonic times.

Twenty years later, the scene was utterly changed. By the interwar period, the historical novel had become déclassé, falling precipitously out of the ranks of serious fiction. There were two body blows to its position in the hierarchy of genres. One was the massacres of the First World War, which stripped the glamour from battles and high politics, discrediting malignant foes and sacrificial heroes alike. Staged by both sides in 1914 as a gigantic historical contest between good and evil, the war left the survivors with a terrible hangover from melodrama. The swashbuckling fare of Weyman or Sabatini looked risible from the trenches. But there was also the critical effect of the rise of modernism, broadly construed, to which Jameson has rightly drawn our attention. He points to its primacy of perception as incompatible with totalising retrospect, rendering impossible a modernist version of the kind of historical novel theorised by Lukács. To this could be added its hostility to the corrupting effects of aesthetic facility – to all that was too readily or immediately available – which struck down the popular and middlebrow versions of the historical novel still more stringently.

Thus if we look at the interwar scene, the historical novel becomes a recessive form, at virtually all levels, in Europe. In the United States, on the other hand, shielded from the shock of the war, Faulkner produced a Gothic variant, flinching before no melodramatic licence, in Absalom, Absalom, while at a less ambitious level its middle range flourished as never before – Thornton Wilder, for example, enjoying a reputation that would have seemed odd in Europe. More spectacularly, Gone with the Wind, a tale of Civil War and Reconstruction with a lightweight resemblance to the romantic nation-building fiction of the previous century, became the most successful historical novel of all time. Significantly, what Europe produced in this middle market mode was principally Robert Graves’s I Claudius, the mental escape of a First World War veteran into antiquity, later fodder for a slack television serial. At a higher level, similar reflexes generated a cluster of historical novels by German exiles – the elder Mann, Döblin, Broch, Brecht – in which Fascism was allegorised into the past, as the rise of Julius Caesar, mobs howling for Augustus, or the killers of the Catholic League, in a deliberately modernising spirit completely at variance with the classical conception of the historical novel.

If this was an enclave with few consequences, two works of the interwar period appear by contrast as signposts to the future. One is perhaps the only work that defies Jameson’s judgment that a modernist historical novel would be impossible, although it is certainly true that it makes no attempt at that ‘sense of a historian’s interpretive commentary on events’ whose absence, he argues, debars Döblin’s Wallenstein – its ferocious canvas of absurdity designed after the senseless slaughter of the First World War – from such a title. This jeu d’esprit was Orlando, whose metamorphoses of time and gender, breaking with every realist norm, occupy a niche in the development of the genre comparable, in its proleptic isolation, to Michael Kohlhaas on the eve of its classical form. Woolf was a modernist par excellence. That an older realist tradition was not extinguished, but still capable of a remarkable reassertion, was shown by Joseph Roth’s Radetzky March, which appeared in 1932. This great novel, which Lukács came to admire, answers to all his criteria save one, which it pointedly reverses. Lukács believed that the true historical novel was carried by a sense of progress, such as had carried Scott. Once this disappeared after 1848, decline to a vitiated antiquarianism set in. The Radetzky March demonstrated the opposite. For Roth’s epic traces the decay of the multinational Habsburg Empire and its dominant class with a clarity and artistry equal, if not superior, to any progressivist 19th-century forebear. A deep historical pessimism proved no bar to a magisterial representation of the totality of objects. Simply, in this reversal the nation-state that was once the ideal horizon of the classical historical novel figures in the novel’s sequel, The Emperor’s Tomb, as the end point of a social and moral collapse – the shrunken, riven Austria of the Depression and the Heimwehr when these works were composed.

Roth’s achievement attracted little notice at the time. The Second World War, when it came, reinforced the effects of the first. The flow of historical fiction at the lowest levels of the genre, reduced but even in Europe never interrupted, swelled again as mass literary markets expanded with the postwar boom: in Britain hoary sagas of doughty patriots battling against Napoleon poured – and still pour – off the presses, from C.S. Forester through Dennis Wheatley to Patrick O’Brian. Over time, this output has yielded a teeming universe that can be glimpsed in such omnibus guides as What Historical Novel Do I Read Next?, with its capsule descriptions of more than 6000 titles, and league tables of the most popular historical periods, favoured geographical settings and, last but not least, ‘top historical characters’ – Henry VIII and Jesus Christ tie for fourth place.

But the larger and more indiscriminate this stratum became, the lower the depths to which the historical novel was consigned as a respectable literary form. In 1951 it came as something of a shock when Marguerite Yourcenar won the Prix Fémina for Mémoires d’Hadrien, so completely out of season did historical fiction of any kind – even such a strange anomaly as this one – seem in the true republic of letters. Did people still write that sort of thing? The profound discredit into which the genre had fallen was made clear by the initial reception of what remains in retrospect the greatest historical novel of the century, Lampedusa’s Leopard. Initially rejected for publication, even when it appeared it was greeted with bafflement by Italian critics. How could such an old-fashioned piece of work have been produced in the contemporary world? Should it be taken seriously as literature?

In fact, what Lampedusa had done was to take the same theme as Roth – the fate of an aristocracy in a dying absolutist order, amid the rise of romantic nationalism – to yet grander conclusions, in a verdict of pitiless detachment on the nation-building process in Italy, the adjustments of the old order in Sicily to it, and the fate of individuals at the crossroads between them, viewed in the light of eternity. Here, the interlocking of historical and existential registers that for Lukács and Jameson defines this form, found supreme expression in the counterpointing of the futile survival of a class and the cosmic extinction of an individual embodying it. Far from being a throwback to Victorian models, the sudden elongation of the novel’s conclusion, jumping 20 years forward to the final disintegration of the taxidermic familiar of the prince, marks The Leopard as a distinctively modern masterpiece.

Strikingly, in the same years that Lampedusa was composing his portrait of the Risorgimento in Palermo, not so far away in the Mediterranean a historical novel was moving in the opposite direction. Naguib Mahfouz’s Cairene Trilogy, depicting semi-colonial Egypt from the rise of the Wafd at the end of the First World War to the activity of Muslim Brothers and Communists during the Second through the story of a bourgeois family, was written under Farouk, and only saw the light of day after the monarchy had been overthrown, in the Sturm und Drang of the nationalisation of the Suez Canal and the Anglo-French-Israeli invasion of Egypt. Mahfouz had begun as a writer of pure costume dramas, Egyptian-style – that is, fanciful romances set in pharaonic antiquity. With the Cairene Trilogy he became a historical novelist such as Lukács has conceived one: real historical figures interwoven with fictional characters, middling heroes, a sense of popular life, and – last but not least – a powerful underlying narrative of progress, however halting or ambiguous, towards national emancipation, even if in this case, technically a Zeitroman, overlapping with his own lifetime. Here it was language rather than prejudice that isolated his work, unknown to the non-Arabophone world till its translation into French many years later.

A reclusive semi-Belgian, a dead Sicilian, an obscure Egyptian. That was about where the historical novel lay, a few antique jewels on a huge mound of trash, for some 30 years after the war. Then, abruptly, the scene changed, in one of the most astonishing transformations in literary history. Today, the historical novel has become, at the upper ranges of fiction, more widespread than it was even at the height of its classical period in the early 19th century. This resurrection has also famously been a mutation. The new forms signal the arrival of the postmodern. To discuss these with due amplitude would require another occasion. The postmodern turn has, of course, extended across virtually all the arts, with local effects distinctive to each of them. But if we consider its morphology in the literary field, there seems little doubt that the most striking single change it has wrought in fiction is the pervasive recasting of it around the past. Since postmodernism was famously defined, by Jameson himself, as the aesthetic regime of an ‘age that has forgotten how to think historically’, the resurrection of the historical novel might seem paradoxical. But this is a second coming with a difference. Now, virtually every rule of the classical canon, as spelled out by Lukács, is flouted or reversed. Among other traits, the historical novel reinvented for postmoderns may freely mix times, combining or interweaving past and present; parade the author within the narrative; take leading historical figures as central rather than marginal characters; propose counterfactuals; strew anachronisms; multiply alternative endings; traffic with apocalyptics. By no means all the historical novels in the vast range produced by accredited writers in the past 30 years exhibit these features. But the core of the revival has typically displayed some or most of them, while around it more traditional forms have proliferated too.

How are we to understand the aetiology of these forms? In a wonderful passage, Jameson speculates that the function of their ‘exaggerated inventions of a fabulous and non-existent past (and future)’ is to ‘rattle at the bars of our extinct sense of history, unsettle the emptiness of our temporal historicity, and try convulsively to reawaken the dormant existential sense of time by way of the strong medicine of lies and impossible fables, the electro-shock of repeated doses of the unreal and the unbelievable’. This is a powerful suggestion. But it raises the question of its possessive pronoun. Who is the ‘we’ of such loss of temporality, that extinction of a sense of history which is ours? Are the postmodern forms of the historical novel effectively universal today?

Certainly, if we were to make a roll-call of all those contemporary novelists who have in one respect or another contributed to the new explosion of invented pasts, the list would stretch around the world, from North America to Europe to Russia to the Subcontinent to Japan to the Caribbean and Latin America. In that sense, such forms have become as global as the postmodern itself. But if we want to track the emergence of the mutation that has produced them, and venture beyond an inventory to their taxonomy, we probably need to consider the spatial organisation of this universe.

No aesthetic timespan is ever homogeneous. The dominance of postmodern forms in the past 30 years did not, and could not, displace all others. At the opposite ends of Asia, something like the classical imagination of the historical novel lived on, producing in Indonesia and Arabia two remarkable cycles of nationalist fiction that can be regarded as, in their way, cousins of Mahfouz: Pramoedya Ananta Toer’s Buru Quartet, composed between 1975 and 1985, and Abdelrahman Munif’s quintet, Cities of Salt, written in the 1980s and already much freer in its handling of time and probability. These are novels, each starting at the turn of the 20th century, written directly out of the experience of Dutch and American imperialism. But they are outliers within the universe of postmodern re-creations of the past. To follow these, we must cross the oceans.

For in point of origin, there is little doubt where meta-historical fiction began. It was born in the Caribbean with Alejo Carpentier’s El reino de este mundo (The Kingdom of This World), which appeared in 1949, followed by his Siglo de las luces (Explosion in a Cathedral) of 1962. Settings: Haiti, Cuba, French Guyana. Five years later came García Márquez’s One Hundred Years of Solitude. Over the next 30 years, Latin American historical fiction became a torrent, with many tributaries beyond Carpentier and García Márquez: Roa Bastos, Carlos Fuentes, João Ubaldo Ribeiro, Fernando del Paso, Mario Vargas Llosa and many more. Here, unquestionably, was the pacemaker for the global diffusion of these forms, which, like the concept of the postmodern itself, were invented in the periphery. Not that sources in the centre were entirely missing from it: Carpentier was steeped in French surrealism; Orlando, translated by Borges, put García Márquez into a fever. But clearly, it was the historical experience of Latin America itself that gave birth to these imaginings of its past. The question is: what experience?

If we set aside individual precursors, the collective take-off of these forms dates from the early 1970s and what they transcribe, essentially, is an experience of defeat: history as what, for all its heroics, lyricism and colour, went wrong in the continent – the discarding of democracies, the crushing of guerrillas, the spread of military tyrannies, the disappearances and tortures, of that period. Hence the centrality of dictator novels in this cluster of writing. The distorted, fantastical shapes of an alternative past, according to this reading, would stem from the thwarted hopes of the present, as so many reflections, admonitions or consolations. It is difficult to deny all force to this diagnosis. But we should remember that the themes of Carpentier’s two originating works themselves, written long before the grim years of continental slaughter and repression, were the Haitian Revolution and the impact of the French Revolution in the Caribbean. These novels, founding texts of magical realism, do not minimise the disappointments and betrayals that overtook each – and which occupy much of the narrative – but their drive is wholly affirmative. The first appeared in the year the Chinese Revolution triumphed; the second just after the Bay of Pigs. Their relationship to the consolidated forms of the fiction they set in motion poses an interesting problem. Could Saramago, a historical novelist whose belated career was ignited by the Portuguese Revolution of 1974, be regarded as a collateral descendant of this now otherwise stranded moment of inception?

In the United States, by contrast, if we consider the span of historical novels of one sort or another produced in the same period, the core experiences triggering the American branch of the phenomenon would appear to be race (Styron, Morrison, Doctorow, Walker) and empire (Vidal, Pynchon, DeLillo, Mailer, Sontag). Here the most distinctive paradigm has been society as conspiracy, not the ostentatious dictator, but the secret network as the hidden ossature of rule: The Crying of Lot 49, Harlot’s Ghost, Gravity’s Rainbow, Underworld – a literature of paranoia offering its own kind of black-magical realism. In Europe, on the other hand, it has been, not the CIA, but the Third Reich and the Judeocide that have polarised historical imagination: Grass, Tournier, Sebald. England, relatively untouched by the Second World War, has generated instead mostly Victoriana – Fowles, Farrell, Ackroyd, Byatt, Carey (an Australian extension) – or reversions to the much more traumatic First World War, as in Pat Barker’s trilogy.

Military tyranny; race murder; omnipresent surveillance; technological war; and programmed genocide. The persistent backdrops to the historical fiction of the postmodern period are at the antipodes of its classical forms. Not the emergence of the nation, but the ravages of empire; not progress as emancipation, but impending or consummated catastrophe. In Joycean terms, history as a nightmare from which we still cannot wake up. But if we look, not at the sources or themes of this literature, but at its forms, Jameson suggests we should reverse the judgment. The postmodern revival, by throwing verisimilitude to the winds, fabricating periods and outraging probabilities, ought rather to be seen as a desperate attempt to waken us to history, in a time when any real sense of it has gone dead.

Still, he concludes, in just these conditions does not the Lukácsian connection between great social events and the existential fate of individuals remain typically out of reach? Benjamin, who detested the idea of progress nurtured by 19th-century historicism, would not have been surprised, or perhaps felt much regret. He used yet another image of awakening. The angel of history is moving away from something he stares at. ‘Where a chain of events appears before us, he sees one single catastrophe, which keeps piling wreckage upon wreckage and hurls it at his feet. The angel would like to stay, awaken the dead and make whole what has been smashed.’ Part of the impulse behind the contemporary historical novel may also lie here.