In the Anti-World

In 1924 the Surrealist Benjamin Péret was eager, like many artists then and since, to relate his own interests to the works of the rich, bizarre and innovative French poet, novelist and playwright Raymond Roussel. In Paris, Péret contacted Roussel’s business manager, hoping to arrange a meeting with the man whom Louis Aragon called ‘the President of the Republic of Dreams’. Members of the Surrealist group, including Péret, had consistently championed Roussel’s expensively disastrous confrontations with good taste in Parisian theatres. At a performance of L’Etoile au front in Paris earlier in 1924, for example, the Surrealists had vociferously applauded while the more conventional members of the (invited) audience jeered, hurling coins and baked apples – cruelly poetic projectiles to throw at actors performing the work of a very well-off man who suffered from anguishing food phobias.

But Roussel shrank from the clutches of any avant-garde clique, insisting, not unreasonably, on his own artistic uniqueness. Péret got seigneurially squashed at long distance, dismissed like a pleading scullion firmly refused entry at the closed doors of the Master’s study. Speaking by telephone from London, Roussel urged his agent to fob off the young poet with the comment that ‘Il ne se classe lui-même dans aucune école.’

Until almost the end of his life Roussel lived in a world that no other artist was either wealthy or neurotic enough to experience. ‘No author has been, or can ever be greater than I,’ he once claimed. And then added: ‘no one is aware of this yet today.’ But this ‘greatness’ was made interesting and anomalous by his writing’s absorption in the everyday, the small, the tangible and the banal. The exquisitely prolonged care that Proust, once Roussel’s near neighbour on the boulevard Malesherbes, lavished on describing the minutest ripples of the interior life are balanced by Roussel’s infinitely patient love of mundane exteriority and the world of objects – dust motes, bubbles, a cork, pencils, a piece of hotel notepaper, grapes, green beans, even the sight of a mildly inflamed uvula. (A distasteful glimpse into the depths of an infected mouth is the nearest that Roussel’s writing ever comes to entering the inner world of another person.)

Mark Ford is alive to the idiosyncratic nature of Roussel’s ‘self-evident uniqueness’ and ‘unassailable self-referentiality’, just as he is aware of the dangers of seeking to contextualise or even to ‘understand’ Roussel’s work, rendering it less freakish or more explicable than it actually is. The man whom Ford describes as ‘the most peculiar of writers’, and whom Fredric Jameson in a phrase of generous understatement has called one of ‘the more aberrant moderns’, seems intransigently committed to an aesthetic of the one-off and anti-real. Roussel’s exceptionally long but tightly constructed works, Ford notes, are inhabited by things such as ‘a bottle-imp containing miniature figurines dramatising a large bird’s attempt to strangle Alexander the Great with a golden thread, or a bas-relief of a girl extracting from a cushion the doll of a one-eyed dwarf dressed entirely in pink, or a skull engraved with Old Norse runes sporting a lady barrister’s cap made out of pages … from the Times’. No other art from the same period insists so calmly and completely on its unassimilablity to conventional canons of the normal, the insightful, the informative, or even the sane.

Eventually and very reluctantly, Roussel came to expect only a ‘little posthumous fame’ for himself. But for most of his life his fantasy of a mass audience and his quest for popular esteem stimulated him to produce a series of extraordinarily complex and original texts. Ford remarks that Roussel believed his writing was ‘destined to convert the masses’, but adds dryly that it has ‘been kept alive only by the interest of successive avant-gardes’. He elaborates on that assertion with plentiful details of the engagement with Roussel’s work of Cocteau, Desnos, Leiris, Soupault, Picabia, Duchamp, Breton, Aragon, Queneau, Perec, Robbe-Grillet, Foucault, Calvino and Ashbery (who contributes a foreword to Ford’s book). ‘I abandoned myself, out of my depth, in the Gulf Steam of your fantasy,’ André Gide wrote to Roussel in 1918 after receiving one of his books, with a sensuous intimacy that Roussel must have hurried to disregard.

Any reader will be able to add further instances to Mark Ford’s impressive list of artists with an interest in Roussel. Take, for instance, Paul Muldoon’s poem ‘The Lass of Aughrim’, which has a character playing a tune on a hollowed-out and drilled human tibia, just as Roussel has in his novel Locus Solus. And the title of Muldoon’s opera libretto, Shining Brow, is drawn from Roussel’s description of himself as one of those gifted souls who carry a star ‘on their shining brow’. A Franco-Irish interchange like this may seem surprising, but that is largely because the current proclivity for embedding authors’ works in sealed and solipsistic national literary traditions makes it that much harder to see just how transnational the whole business of writing then was and still remains. It was Flaubert who said: ‘I am as much Chinese as French.’

Like Gide, Ford allows himself to float trustingly in the immense, meandering, warm(ish) current of Roussel’s writing. But he does so while remaining resourceful in his marshalling of scholarly sources, passionate in his advocacy, and cosmopolitan in his awareness of avant-garde art in several media. Raymond Roussel and the Republic of Dreams is a lucid and often witty volume, wholly worthy of its subject and of absorbing interest both to Roussel aficionados and to those whom Ford envisages as having ‘little or no acquaintance with Roussel’s oeuvre’. Written about an artist who did not ‘class himself as belonging to any school’, it is, aptly enough (and not only because it is the first analysis in English of Roussel and his work since Rayner Heppenstall’s short and now wholly superseded 1966 volume), a book in a class of its own.

Roussel was born in Paris in 1877, the year Edison patented the phonograph, and he died from a drug overdose in 1933, at the height of a Europe-wide yo-yo craze, killing himself in the Palermo hotel where half a century earlier Wagner, one of his artistic heroes, had finished Parsifal. (The hotel’s website currently boasts of being ‘a home-from-home for many illustrious guests’, and mentions Wagner’s triumph, though it does not indicate that Roussel committed suicide in one of the ‘183 rooms’ – the one that is today number 225.)

He was the third child of a colourless speculator in Paris’s booming real estate market and of an independently wealthy mother. Even in hard times (so to speak) in the 1920s, Roussel’s estate in Neuilly, bequeathed to him by his mother, had a staff of 16, including a chef, a governess – he had no children – a maître d’hôtel, two foot servants, a washerwoman, three chauffeurs, three gardeners, three cooks and a valet de chambre. There, a world of dreamlike, amoral, atemporal artifice prevailed. Roussel would order that the entire day’s meals, prepared by an Escoffier-trained cook, be served to him in what Ford calls a ‘single marathon sitting’. At 12.30 in the afternoon a lavish breakfast began, followed immediately around 1.15 by an extensive lunch, succeeded promptly by an early but immodest dinner. Each day Roussel rose from this mighty repast at about 5.30 p.m. with – on account of his eating difficulties – the board still groaning as it had been five hours earlier.

Such pathos-filled extravagance was possible only because Roussel was easily the wealthiest serious author of the 20th century. In fact, delete ‘serious’. He was surely the period’s wealthiest author. In comparison, the circumstances of the very set-up indeed James Merrill (like Roussel an addicted poetic formalist), driving his own Volkswagen and using a homemade Ouija board all those years, seem positively peasant-like and self-reliant. Alison Lurie recently described Merrill in Connecticut, chatting in his kitchen as he sliced up ‘a ripe red tomato with his little serrated knife’. Roussel, by contrast, did not want any indication that the food on his plate in the carefully controlled world of Neuilly had been violated by anyone at all. Forswearing any equivalent to Merrill’s ‘tiny serrated knife’, he returned belowstairs any produce with even the faintest trace of an incision.

The eerie, arbitrarily imposed rules and rituals of Roussel’s adult life were perhaps to be expected from someone who had grown up in a home consecrated to a chilly rigmarole of fancy and play. His mother was an avid collector of faience and fans, porcelain, jewels, Old Master paintings and lapdogs – her Biarritz villa contained minibaths for her brood of chihuahuas. And Mme Roussel amassed other small things besides dogs: she owned a significant collection of Dresden figurines; 78 were in her possession at the time of her death in 1911. Moving among this miniature army of bloodless shepherdesses and porcelain cattle, she inhabited a realm of whimsical petrifaction; one which must have had a formative influence on her favourite son, who grew up to be the creator of some of the strangest scenes and most intricately stylised artefacts in modern literature.

Where Proust spent a lifetime excavating the caverns of childhood memory, Roussel had no inclination to make a representation of his own inner life in words. His travel jottings from the important journey he made to Egypt in 1906 belong to a different verbal universe from Flaubert’s wonderful writing about his trip to that country in 1849-51. Almost as if he were performing a technical experiment in total self-suppression, Roussel makes no mention whatever of his reactions (‘Went to see the Valley of the Kings – Cold lunch – sun – heat’), even though his last and greatest poem, Nouvelles Impressions d’Afrique, published in 1932, begins with the provoking sight of an Egyptian door. A few such objects aside, the author of Impressions d’Afrique and Nouvelles Impressions d’Afrique seems to have had no impressions of Africa. Which was the point. ‘Chez moi, l’imagination est tout,’ he exulted.

He began writing poetry at the age of 17, in the same year as his father died (in a swimming-pool after, Roussel’s biographer François Caradec reports, drinking iced champagne on a hot day). Two years later he wrote La Doublure, a 5600-line poem which until the end of his life he continued to believe was one of his greatest achievements. It was the first of his works to appear in print (at his own expense, like everything else he wrote). While composing La Doublure, Roussel felt suffused by a total happiness, ‘la gloire’: ‘I was carrying the sun within myself and could do nothing to impede the tremendous light I was radiating.’ And, believing he was writing ‘with a thousand flaming pen-nibs’, Roussel was certain that fame (also in French la gloire) would be his: ‘I was the equal of Dante and of Shakespeare, I was feeling what Victor Hugo had felt when he was 70, what Napoleon had felt in 1811 and what Tannhäuser had felt while musing on Venusberg.’

These recollections were confided to Pierre Janet, the psychiatrist who wrote up Roussel’s case in From Anxiety to Ecstasy (1926), a volume on religious manias. Janet chose for his patient the pseudonym ‘Martial’, the first name of the central figure in Roussel’s novel Locus Solus, published in 1914, at around the time when Roussel began to consult him. Janet’s lengthy (and doubtless expensive) attentions prompt the thought that Roussel, so deeply convinced of his own unique worth, had one of the longest-lasting and most confiding relationships of his adult existence with a person who openly considered him psychotic. His only comparably close liaison was with his paid companion (another expensive necessity, this time occasioned by the beau monde’s prejudice against homosexuality), Charlotte Dufrène, whom he was introduced to in 1910, and who faithfully accompanied him right up to the terrible trip to Palermo 23 years later. ‘Better go down dignified/ With boughten friendship at your side/Than none at all,’ Robert Frost wrote. Roussel’s boughten friendships with Janet and Dufrène explored the feasibility and not unsuccessful consequences of Frost’s idea.

When La Doublure appeared it received almost no attention. The second review, published five months after the poem’s appearance, called it ‘very boring’, a verdict evocative of Jameson’s suggestive remark a century later: ‘it is no great secret that in some of the most significant works of High Modernism, what is boring can often be very interesting indeed, and vice versa.’ He instances Roussel to push home his claim. After the non-event of La Doublure Roussel wrote that he ‘plummeted to earth’ like Icarus, ‘from the prodigious heights of glory’. From then on, as if seeking to replace effortless inspiration with concentrated will-power, he wrote assiduously, even when his efforts issued only in intense frustration. At least two periods of several years were spent ‘prospecting’: searching, that is, for a method and an artistic way. The wealthiest of individuals, Roussel toiled at his writing like a hod-carrier. He commented on his ‘fever of work’ and, while writing his novel Impressions d’Afrique, told Janet: ‘I bleed over every phrase.’ When Michel Leiris saw Roussel for the last time in 1933, he asked whether he was still writing. Without exactly saying yes, the ashen poet replied affirmatively: ‘C’est tellement difficile!’

As a result of this regime, and without ever having to submit to the indignities of the literary marketplace, Roussel became the author of a large quantity of high-level work: two complicated novels, Impressions d’Afrique and Locus Solus, multiple volumes of poetry, from 1911 on several adaptations of his work for the stage, and two original plays, L’Etoile au front and La Poussière de soleils. The extent of Roussel’s productivity became still more impressive after the discovery in 1989 in a Paris warehouse of a trunk containing ‘thousands of pages of rough drafts, fair copies and proofs’ as well as a huge, previously unknown verse play, La Seine, of seven thousand lines with speaking parts for over five hundred characters, and Les Noces, another unstageable drama in three parts extending to over twenty thousand lines.

Roussel’s arduous and (from an economic point of view) completely unnecessary work routines were accompanied by other costly forms of self-disciplining. Eating, as we have seen, was a particular problem. Leiris reports that after fasting or consuming little while he was working, Roussel would go on blow-outs at expensive restaurants, indulging his ‘taste for childish foods: marshmallows, milk, bread pudding, racahout’. Details like these add emotional colour to the many memorable food moments in Roussel’s work. There is, for instance, the life-size statue of the Spartan slave, a helot, clutching at a sword thrust into his heart in Impressions d’Afrique. The Spartan is made from black whalebones and rides on a whalebone trolley, the wheels of which run on red rails constructed from a gelatinous preparation of calves’ lungs. (The very choice of a dying helot to be suspended on tracks made from animal organs is itself knowledgably and sensitively foody: Xenophon claimed that the helots would have liked to eat the Spartans raw.) In fact, a simultaneous fascination and disgust with food is evident throughout Roussel’s writing. A little later in Impressions d’Afrique we meet a sculptor who has invented a pastille ‘capable of shaping water into momentary reliefs’. The level of detail in these drawings is so hysterically sharp that one can even notice ‘the shadows cast by breadcrumbs on the tablecloth’.

And with fastidiousness about food (what is allowed into the body) went an obsession with dirt (tell-tale evidence of what is allowed out of the body and of its interactions with the world). Roussel spent as much as two hours at his dressing-table before going out in the evenings, and would wear a collar once, ties three times, and a suit or overcoat 15 times. His companion Dufrène was allowed to wear a pair of white gloves only once before disposing of them. Ford relates that during a stay at the Ritz in 1928, Roussel was sending as many as 25 handkerchiefs a week to the laundry.

But Roussel’s crazy life, like his work, was all about suggestive tensions and paradoxes. This man sedulously cocooned inside the anti-world that his unlimited wealth had spun around him was also a dedicated traveller, mainly to continents beyond Europe. He went to Egypt in 1906, to Ceylon and India in 1910, and then made a round the world voyage in 1920-21 with stops in India, Australia, New Zealand, China, Japan, the US and Tahiti. In Polynesia he went in search of the originals for his literary god Pierre Loti’s sybaritic fictions, but claimed to have missed the tropical sunsets because he was busy working in a shuttered cabin.

Around 1924, Roussel designed and had constructed a camper van complete with bed and bathrooms, a small dormitory for servants, a safe and a ‘radio capable of receiving all the major European stations’. It was exhibited at the 1925 Salon de l’auto in Paris, where it drew discriminating attention. The vehicle was called a roulotte, a fact which must have delighted its owner, fascinated as he was by chance recombinations of language as well as by the sound of his own name. Early on he had coined roussellâtres, to designate his own imaginary acolytes, and he delighted in sending his books to individuals called Blanche Roussel, Henri Rousselle, and even Raymond Roussel (who received a copy of Locus Solus inscribed ‘to Raymond Roussel, presented by his homonym’). That Roussel’s roulotte, in compliance with the roulette-like operations of verbal chance, should verbally embody a tiny chunk of his own name must have been a special treat.

He took his camper van to Italy in 1926, where it was admired by the Pope and by Mussolini, both of whom accepted commemoratory photographs of this forward-looking contraption. In this case, as so often, Roussel’s seemingly unworldly eccentricity turned out to be closely attuned to the historical moment: in his aristocratic way, he had stumbled on the populist idea of the motor home. The first American example had been built as early as 1915 and by the mid-1920s, when the roulotte was rolled out, two more custom-built vehicles were in use in the United States. But the roulotte was to be the last of Roussel’s prescient and creative follies. His return from a long trip to the Near and Middle East in 1927, including stops in Turkey, Palestine, Syria, Iraq and Iran, coincided with a period of steep personal and financial decline, the accumulated toll of so many theatres and actors hired to put on intrinsically unperformable dramas, of blackmailers paid off, of books seen into print at vastly inflated cost. He sold the Neuilly estate, and eventually moved into an apartment in his sister’s Paris house. Now in the final stages of composing and recomposing Nouvelles Impressions d’Afrique, Roussel began to drink heavily. And, as his struggles with this most complex and significant of all his books became ever more wearying, he began to take large quantities of barbiturates, searching for the blissful sensations of self-completion and omnipotence he had experienced long before, while writing La Doublure.

In 1932, as his life darkened still further, he developed a new passion – for chess. Naive as this strange life-artist may have been, even in his harried final years he retained a remarkable intelligence. In spite of being deep into his own bodily endgame, he rapidly worked out the ‘Formule Raymond-Roussel’, which he described as a ‘means of achieving the extremely difficult checkmate with bishop and knight’. The formula was taken seriously enough by chess professionals to be written up in magazines.

By 1933 Roussel had apparently come to terms with his own imminent somatic checkmating. In January, he deposited his literary testimony, How I Wrote Certain of My Books, with his lawyer (ordering it to be sent after his death to the Surrealists he had earlier spurned) and in early June he and Dufrène left for Palermo, where he spent afternoons being driven round the city and nights taking pills. But, rather than compulsive ingestion, the motif of his final days was to be the continual spilling of precious fluids, a kind of non-urinary incontinence. Finding unconscious the writer who had once described a cock coughing up letters of blood, the hotel doctor performed a blood-letting. At the start of July, after consuming a bottle of Neurinase, Roussel promised the hotel valet, Orlando, a bribe if he would slit his wrists for him. The valet refused, but the next morning Roussel was found in his bath with slashes at his left wrist and throat. ‘How easy it is to open one’s veins,’ murmured the man who had once said of his writing that ‘I bleed over every phrase’ and could not bear to see a knife-mark on a vegetable.

Patched up, he kept offering Dufrène money to shoot him. Ford elegantly remarks: ‘Money had failed to confer on his work the acclaim that might have made his life bearable, and it now failed in his irresolute search for a means of ending his existence.’ During the early morning of 14 July 1933, Roussel died alone in his room, probably from the combined effects of a lack of food, alcoholism and an overdose. The police found a ‘pale, scarred corpse’ lying on a mattress on the floor. Near the body were 12 copies of Locus Solus, the central chapter of which describes the scientist genius Martial Canterel performing experiments during which eight cadavers, under the influence of ‘resurrectine’ and ‘vitalium’, are reanimated and ‘compelled to re-enact the most crucial moments of their lives’. Orlando, who had earlier declined to slit Roussel’s wrists, saw fit to note for the local Fascist authorities that in his death agony Roussel had ejaculated. (The hotel, according to its website, is still ‘offering the best of service to discerning visitors to Palermo since 1874’.)

Perhaps these are trivial matters; except that they were not for Roussel. A person who had believed for so long in his own literary immortality was in the end fearful not so much of dying as of not really being dead. Before leaving Paris, Roussel had left instructions with his lawyer: ‘I absolutely insist that a long incision is made in the vein of my wrists so there is no danger of my being buried alive.’ His remains were interred at Père Lachaise in a plot which contains spaces for 32 bodies, the number of squares that, once the game is over and someone has won, someone lost, are visible on a folded chessboard.

Written in a French that one commentator described as ‘clear, pure, precise and unflinching’, and inflexibly deferential to the spectrum of Neoclassical stylistic protocols, Roussel’s books search for a state of immobilised poise. This goal of suspendedness is clear in poems such as La Vue, where two thousand lines describe what Roussel sees when his ‘wide-open eye is glued/Very close’ to a souvenir pen-holder containing a minuscule photograph of a beach scene. His writing, as Ford remarks, is endowed with ‘a kind of Midas touch: everything is metamorphosed into aesthetic pattern, death as well as life.’

It is logical then that in Roussel’s work the animate and the inanimate should approach from opposite directions some artistic middle ground beyond the binaries of life and death, human and inhuman, male and female, perhaps even rich and poor. On that terrain human beings seem less than alive, while objects and animals apparently acquire human capacities. In Impressions d’Afrique, for instance, the six Alcott brothers, all of an ‘extraordinary thinness’, take up positions in a garden. When their father calls out, each of the young men reverberates to produce arpeggios (‘they exposed to the sound a surface of bone, hard enough to reflect all its vibrations’). Humans thus become instruments while elsewhere insects become musicians: a giant worm secretes a special liquid onto a zither to produce pleasing tunes.

Shapely, emotionless fantasies like these make it plain that Roussel’s texts aren’t intended as descriptions of the world or of recognisable states of mind. But his anti-character, anti-anthropocentric bent is most prominent not so much in his subject-matter as in the aesthetic forms that his work takes. Everything he created aspires to the condition of a silent, self-enclosed, seemingly unbreachable structure, sealed for ever like a Victorian mausoleum. In this respect, Roussel’s fascination with aesthetic construction and exteriority rather than self-expression emerges from a major tradition in 19th-century French verse. Baudelaire repeatedly insisted that poetry was ‘an object’ and in the work of many of the period’s writers, including Baudelaire himself, the figure for that ‘objectness’ is often a statue: a free-standing, hard, indifferent form, like a Spartan made of whalebone, lifeless for all its verisimilitude, and positioned in absolute contrast to the volatile neediness of the desiring human mind which observes it. Baudelaire’s sphinx exults sadistically: ‘I am beautiful, oh mortals, like a dream of stone.’

But it is precisely the ‘dream of stone’, the lapidary independence and unchangingness, that is so stirring for the poet and that becomes the ideal to be re-created in language. The motif is taken up by many of the later Parnassians, poets whom Roussel nostalgically endorsed, to the great bewilderment of the otherwise unshockable Surrealists. Throughout his career he insisted, as if to suggest genius by association, that his writing be brought into print at his own ruinous expense by the deeply unfashionable publishing house of Alphonse Lemerre, the original publisher of the anthologies Le Parnasse contemporain in the 1860s and 1870s and subsequently of many of these once prestigious and now forgotten versifiers.

Ford offers a brilliant account of the autonomous structures of Roussel’s work, emphasising the gap between text and context, art and life. ‘It is the uncanny distance the work establishes between writer and reader’ that is for him the source of Roussel’s strange power. His analysis develops by returning again and again to the ‘patterning and symmetries created by Roussel’s works’. Indeed, ‘antithesis’, ‘pattern’, ‘symmetry’ – these terms surface, vanish and resurface in the book like the recurring endwords in a sestina or the symbols in a dream. And in this way, Ford’s book seems representative of the taste for ‘form’ shared by some of today’s best poets and poetry critics – not so much in the sense of writing composed in relatively predictable, traditional shapes such as the villanelle or ballad, as for half-hidden, intricate and enigmatic systems of verbal organisation proliferating within and between poems. This is, for instance, a characteristic preoccupation both of Helen Vendler’s recent study of Shakespeare’s Sonnets and of Paul Muldoon’s collection Hay.

Much of early 20th-century avant-garde practice was dedicated to breaking down rules; Roussel by contrast was dedicated to a fanatical but arbitrary imposition of constraints. And his self-imposed rules centred on the quasi-independent machinations of language. He was obsessed with linguistic instability and volatility – the kind of doubleness visible in puns, which point semantically in two directions at once – and with finding a means to overcome this nightmarish slipperiness. Puns, in fact, were the fundamental factor in Roussel’s ‘very special method’ (the ‘procédé’), which he evolved in his early twenties as a way to generate in prose ‘unforeseen creation due to phonic combinations’. In subsequent years he progressively refined the ‘procédé’ by making it an even more capricious and independent device.

The most famous example of the ‘procédé’ (Roussel gives it pride of place in How I Wrote Certain of My Books) comes from an early story, ‘Parmi les Noirs’, which begins with ‘Les lettres du blanc sur les bandes du vieux billard’ (‘The letters’ – of the alphabet – ‘in white’ chalk ‘on the cushions of the old billiard table’) and ends with ‘les lettres du blanc sur les bandes du vieux pillard’ (‘the letters’ – i.e. epistles – ‘sent by the white man about the hordes of the old plunderer’). With these two phrases in place, and distinguished from each other only by the typographical flipping of a ‘b’ into a ‘p’, Roussel asserted that ‘it was a question of writing a story that could begin with the first and end with the second.’

The procédé works by generating an initial uncertainty (what does this mean? what will/must happen next?) which it then becomes the task of the ensuing text to resolve by getting as close as possible to the original starting point. As Ford says, the ‘procédé opens a fissure in language which the resulting story must then bridge’. Faced with this device, one might be tempted to rewrite Axël’s remark – ‘As for writing, our language will do that for us.’ But actually, as Roussel’s eerie and ingenious texts demonstrate, it takes extraordinary amounts of hard work, calculation and compositional inventiveness to effect this resolution.

At one level the procédé, continuously deployed, is a parody of the mechanical ‘formulae’ that popular writers of the period, such as Conan Doyle or Loti, were using to generate endless sequels to an initial lucrative success. At another, it is a wholly serious attempt to displace attention from a work’s content to its style and shape. Either way, this interest in the anarchic, internally generative potential of language marks Roussel as a creature of the late 19th and early 20th centuries when all kinds of anxiously ludic homages to the autonomous powers of the word were springing up. Jules Verne, one of Roussel’s favourite writers, secreted verbal enigmas throughout his writing, as did Poe, another influence; the period before the First World War also saw the emergence of such disparate but related phenomena as the Spoonerism, synthetic tongues such as Esperanto and Ido, and the crossword.

The procédé was a machine to make prose with, but Roussel found analogous methods to activate his verse. Ford tells us that the recently discovered manuscripts show Roussel using ‘rhyme as the crucial first stage in originating his poetry’. (Théodore de Banville, a fellow Lemerre author, had asserted in his Short Treatise on French Poetry, 1872, that rhyme is the essence of poetry: ‘rhyme is the only harmony of verses and it is the whole verse.’) The effort to achieve perfection in verse proved just as arduous as the effort in prose: Nouvelles Impressions d’Afrique is only 1274 lines long yet, such are its procedural complexities, Roussel was stuck in its labyrinths for some fifteen years – an overall rate of progress equivalent to about a line and a half a week.

In Nouvelles Impressions the governing principle is lateralism, a concept that is given concrete form by Roussel’s remorseless deployment of the parenthesis, which involves continual operations of sidestepping (just like the moves in chess of the bishop and the knight, the two pieces used by Roussel in his ‘extremely difficult checkmate’). The poem begins in Egypt, the land ‘of crumbling marvels’. A door that led to the prison where Louis IX was once briefly held captive suggests a ‘thought’ about how in comparison with the ancient Egyptian buildings everything else seems new. This introduces a thought about familiarity, and this introduces a thought, encased in a parenthesis, about other kinds of familiarity. But nested inside that parenthesis is another digression on the art of photographic retouching, and within that another on how people respond to having their picture taken. Within 12 lines of opening his poem’s first parenthesis, then, Roussel is immured within five levels of bracket: ‘Le fat qui d’ un regard (((((parfois une étincelle …’ And inside those five brackets, Roussel complicates the maze still further by dropping a footnote – like all the notes in the poem, it is in rhyming alexandrines – at the bottom of the page.

There are so many tangents and zigzags through which the line of the poem passes that, between the time when the first parenthesis opens in Part 1 and the time when it closes, the path from its beginning to its end has been separated and interrupted by four succeeding parentheses (going ‘sideways’), several footnotes (enforcing movement up and down the page), ten beautifully inert illustrations (specially commissioned by Roussel, through the veil of anonymity provided by the mediation of a detective agency, from a modestly talented illustrator), some 215 lines of loosely connected poetry (roughly a sixth of the poem’s entire length), and, for its readers, endless shuttles back and forth between pages to remind themselves of what the sentence they are now suddenly reading again was saying before it was interrupted in mid-flow by a lengthy sequence of bracketed digressions.

Roussel’s last poem brings a desire for control and order, for a culminating demonstration of all-mastering technical and technocratic supremacy, to the centre of the reading experience. And that control is as much over the progress of the alternately exasperated and beguiled reader, swerving, backtracking and skipping forwards (all at the Master’s inscrutable whim), as it is over the ‘irreproachable’ language of the poem itself.

Ford admits that the necessity of holding six uncompleted lines of thought in mind at the same time involves ‘formidable difficulties’ for the reader (‘form-idable’ is a nice pun in this context). But he argues that ‘if the brackets continually rupture the progress of a particular line of argument, they also serve as forms of connection, almost as a synapse or railway points that enable the poem to cross into whole new regions of proliferative analogy and illustration.’ This gestures acutely to the huge paradox on which Nouvelles Impressions d’Afrique rests, and from which it draws its enduring interest. The rigorous, almost manic assault on communication, realism, sequence, reference and predictability which the poem launches – apparently to promote awareness of its own unique, independent verbal architecture – coexists with an incessant tendency to collapse back into the social and historical world it longs to lock out. Roussel’s final piece of completed writing is poignantly unable to retreat into a hermetic sphere of pure, undifferentiated artfulness – a ‘failure’ that is the key to its weird success.

In a strikingly harsh phrase, Georg Lukács referred to Faust as a poem of ‘capital running with blood’. Something similar could be said of Roussel’s last major piece of writing. (He carefully mentions a detail from Goethe’s epic in Nouvelles Impressions.) Even at its most intricately formal, it is provocatively awash with uncontained liquids which pour around its formal boundaries like a tide sweeping over a harbour wall. Seepings, spills, drops and emissions, often of human origin, burst out from a hundred verbal gullies in Nouvelles Impressions. A parallel with Ulysses, published in Paris in 1922 when Roussel was about halfway through the composition of his poem, is suggestive. Joyce’s novel, which he repeatedly called an ‘epic of the body’, is also crucially laced with emissions, sloshings, spillages and liquid expulsions. For Joyce, the streams are among other things a way of getting beyond the confines of literary realism. For Roussel, the leaks are expressive figures for the impossibility of creating a perfect verbal container which can impregnably insulate its content from contact with the world.

Ford notes that Roussel’s poem is replete with ‘bodily discharges, impurities and waste products’. Roussel’s extremely well-wrought urns always have strangely anthropomorphic holes, fissures and cracks in them: like the ‘Crow’s feet and spots’ that a clever photographic retoucher knows how to remove from a portrait, or the pus running from a blister punctured by a pin-prick, which is compared with water ‘running away in the desert/Which the sword of a traitor released from a goat-skin’. Elsewhere in the poem we encounter sheets stained with menstrual blood, bloodstained handkerchiefs, horse droppings, a goat defecating, a bottom being wiped with a sheet of newspaper (just as Bloom wipes himself with a torn-off page of Titbits in the Calypso section of Ulysses), though unlike the material that Roussel had earlier described as being used for the lady barrister’s cap he makes clear that the paper in question here is not the Times, a shirt-tail about to be stained by a heavy flow of diarrhoea, a bird thinking of shitting on a picnic, a drunkard’s strong clear ‘horizontal jet of urine’ (reminiscent of the double streams in the Nausicaa section of Ulysses), a fragrant flower that has been pissed on. This epic of formal closure and self-involution is permeated by a cascade of unstaunchable leaks that makes Nouvelles Impressions look at times like the verbal equivalent of a colander. (Is an obsession with ‘flow’ an understudied general feature of Modernism, perhaps? Besides the copious similarities in Ulysses, compare Eliot’s ‘liquid siftings’ and his rubbish-filled Thames and Mississippi or indeed Beckett’s poem, written in Paris the year after Roussel’s book appeared there, with the title ‘Sanies’, meaning ‘a thin fetid pus mixed with serum or blood, secreted by a wound or ulcer’.)

This foul symphony of leaks does not simply suggest an autobiographical fear of bodily dissolution, though that is surely relevant to a poem composed in large part when Roussel was worrying about his own physical decrepitude. The dripping breaches also stand for the poem’s triumphant porousness, its inability to close itself off from the world. Ford is right to suggest that the brackets, ostensibly used to allow the poem to interrupt a thought and head off in another direction, are also forms of ‘connection’. For another marked feature of the poem is the enormously diverse range of scenes and people which the poetics of analogy can draw together. The effect is not to drive readers into a non-referential linguistic morass, but to impel them insistently back into the historical world of events, objects, interactions, with an obscure sense that these multiple and apparently disparate circumstances are all in some way interrelated even when they are separated by vast distances of place and huge discrepancies of social scale.

It is no accident, then, that by focusing on the door behind which Louis IX was briefly held prisoner the poem begins by invoking a sense of ‘history’. The weblike universe of Nouvelles Impressions d’Afrique spans Western and non-Western worlds, from the cell-like precincts of a manned elevator in a Paris apartment building to the huge, empty spaces of the African desert. In this sense, Roussel’s relatively short but exceedingly complex poem offers a microscopic image of a huge, ragged, but systematised society, spread out panoramically for contemplation much like the endlessly copious scene unfolding in the minuscule ambit of the photograph inside the pen-holder that Roussel had gazed at almost three decades earlier. (He explained posthumously that the new poem had been meant as ‘a follow-on to my poem La Vue’.) The connections often remain frustratingly hard to fathom and difficult to remember but taken together they offer themselves as an abstract image of any system so vast and complex as to be only partially assimilable and comprehensible: the system of a large, sophisticated machine perhaps, or a modern city, or, maybe – the poem’s point of departure in each of its four cantos is Egypt – of a colonial empire. For all its apparent self-designed seclusion from the world, Roussel’s poem may well be what Adorno said all lyric poetry was: ‘a philosophical sundial telling the time of history’.

The idea that a recognition of the shared, historical world is at the heart of this ostentatiously ‘formal’ poem is deepened by the mundaneness of its subject-matter and language. Triteness and banality are two important areas of exploration in early 20th-century poetry, as the interests of both Hardy and Eliot in the poetry of the commonplace, the tawdry and the unremarkable demonstrate. But Roussel’s resourceful dullness of sentiment – existing in a relation of mutually dependent contrast with the poem’s amazingly planned, intricate form – is even by these benchmarks impressive. Ford picks out a few amusing examples of the ‘series of truisms’ with which the poem’s analogy-making machinery is kicked into gear: that, contrasted with the ruins of Egypt, even old things seem new, that we are all sometimes in need of help, that one can get used to anything, that the form of the cross can be seen in many different contexts. Again, the poem generates a strange paradox – without being vacuous, it is memorably empty.

But I am making the poem, or rather the experience of reading the poem, sound like a more dispassionate, ‘literary’ affair than it actually is. It is striking that authoritative Rousselians such as Ford or John Ashbery, one of Roussel’s earliest and still most unflagging supporters, seem to describe his work with a beguiling sense of calmness and equanimity, in a tone a bit like that of a pathologist describing a wound to a colleague. In a wonderfully non-judgmental sentence Ford writes that ‘La Doublure seems devoid of an ascertainable artistic intention; one searches in vain beneath its painstakingly detailed surfaces for clues to its ulterior purpose.’ To me Roussel seems more interested in the cultivation of palpable, and sometimes searing, moods and reactions in his readers than this would suggest. Just as in Nouvelles Impressions the form’s rule-bound autonomy paradoxically returns the reader to the world of history, so the poem’s studious emotional neutrality coexists with a sense in which this work is deeply ‘affecting’. I find that its structural difficulty, its recessiveness, its caprices can, as with much of Roussel’s work, drive you to the point of distraction or delight. I hadn’t read Nouvelles Impressions d’Afrique before, and it took me slightly over a fitful month to read its 1274 lines. It was a month in which I many times gave up on the poem, and many times went back to it, a month in which while I was reading my mood jumped around like a monkey in a cage from amused to bored, delighted, bewildered, interested, furious, cynical, absorbed, elated, deflated, a month in which I felt intensely aware of my life slipping away from me, a month in which I experienced both admiration and pity – for Roussel and, perhaps more surprisingly, for myself. What Roussel’s poem offers is not just a Stendhalian ‘promesse de bonheur’ but a plenitudinous ‘promesse de malheur’ as well.

The poem’s mawkish beginning (‘behind that door,/The saintly king was for three months a prisoner! … Louis the Ninth!’) unavoidably ignites an initial mood of supercilious interest. The reader seems to have power to spare over the poem. But slowly and inexorably, as the complexities of the form take hold, as the impossibility of reading the poem in a sequential fashion becomes apparent, as the strange interlacing of scenes begins to suggest a vast, perceptible but incomprehensible order, the balance of power tilts decisively towards the poem and away from the reader. And as it does the reader becomes aware that, like so many of the great works of Modernism, Nouvelles Impressions may not be about life’s finer feelings, but it is about life’s feelings (you know this because you start to feel them all) – panic, rapture, boredom, amusement. Should writing give its readers only positive sensations like pleasure? Ford writes that ‘the sovereign I of la gloire, who believed he was about to illuminate the entire universe, undergoes a correspondingly absolute annihilation, his lyric self wholly diffused amid the poem’s multiple, parallel lines of discourse.’ But this is not just an experience that the author undergoes; the reader does, too, in ways that are sometimes painful and sometimes, well, mind-expanding. You have to abandon yourself to the Gulf Stream of Roussel’s fantasy, as Gide claimed he had done, and the process is a truly novel and disconcerting experience, a deregulation of all the senses.

I cannot see myself going back to this poem for a while (it is too exhausting, too intense for that), but I will not want to forget it either. Nouvelles Impressions d’Afrique is a limit-text ranking in its experimental strangeness and implacability with many other creatively extreme books of the 1920s and 1930s. These works have now been around for a whole human lifetime or more. And yet, as a reading of this remarkable poem suggests and as Mark Ford’s superb volume on Roussel and his continuing artistic aftermath confirms, there are times when it can seem as if the clichés of literary history tell the truth (on disinterested as well as self-interested grounds Roussel himself would surely have believed they do), that in a new century we are still living under the sway of those inhuman masterpieces, still ruminating in their wake.