Paradise Regained: Queen Kelly and the Lure of the ‘Lost’ Film Darragh O’Donoghue July 2003 Lost Films Issue 27 When we talk about ‘lost’ films, we are probably referring to a few masterpieces that were taken from some Great Director and reworked by hacks for various, but usually base and commercial, reasons – e.g. Greed (Erich Von Stroheim, 1925), The Magnificent Ambersons (Orson Welles, 1942), The Red Badge of Courage (John Huston, 1951): noble travesties that can never be repaired because the edited footage was destroyed. This concept of the lost film is indispensable to the great Romantic myth of the cinema, the endless struggle between Art and Mammon, with the visionary director cast as David against the Philistine might of the studio system, but with none of the young Judean’s power to win through. We may bewail the crass motives that forever maimed these masterworks, and the experience of watching them can be painful and frustrating, but, if we’re honest, we’re secretly glad: if Ambersons, say, did survive, it would be just another classic for us to sit back and admire, not the unique treasure it now is because, as David Thomson suggests, it is lost. The lost film speaks to an ambivalence at the heart of cinephilia. There is, perhaps, a sense that the cinema shows too much, places the viewer in too passive a position; the lost film, the fragment, the butchered stump, so tragic a loss of an original vision, nevertheless liberates the viewer, activates his or her imagination, to take what remains, and, unbounded by human or historical probability, send it in directions the filmmaker may never have intended. This applies especially to a peculiar form of ‘lost’ film, the project mooted but, for whatever reason, never executed – Albert Lewin’s adaptation of Oscar Wilde’s The Picture of Dorian Gray with Greta Garbo in the title role; Kubrick’s Napoleon project; or those famous follies that foundered on production difficulties, such as Sternberg’s I, Claudius (1937) and Welles’ Don Quixote (1959). You might even extend the fancy to those projects you imagine – an adaptation of some favourite book, casting as you read; the unexpected pairing of incongruous talents; or those famous films that almost starred someone else (Ronald Reagan in Casablanca [Michael Curtiz, 1942]; Anna Karina in À bout de souffle [Jean-Luc Godard, 1960]; Paul Newman in La Dolce Vita [Federico Fellini, 1960]. All this might point to a tacit disappointment with the cinema as we know it and a yearning for the Platonic ideal we dream it capable of. Perhaps the lost film reclaims some human fallibility for an artform that is also a reproductive technology – its initial promise to record transient lives and places for all time betrayed by its own vulnerability to erosion. The tension between the fragmented and the complete harks back to the early years of cinema, and the classic dichotomy of the Lumières and Méliès, the one offering snippets of life en plein air wherever a camera could be placed, the other finished fantasies in the same studio – the former, to many, still seem fresh and strange long after the endless tricks of the latter have exhausted. And it is an ambivalence recorded in the critical appreciation of film – Citizen Kane (1941), Welles’ one unmolested production, is repeatedly ranked in polls above equally extraordinary, but broken masterpieces like Ambersons and Touch of Evil (1958). Its partner at the top, however, is usually Jean Renoir’s Le Règle du jeu (1939), long considered lost after the depredations of war (an industrious industrialist of lost films), and surviving today as a patchwork of fragments artfully arranged to give an illusion of wholeness. We forget, of course, that in its earliest decades, film was not generally considered an art worth preserving, and so the vast majority of silent films – including countless potential masterpieces by the likes of Feuillade, Vidor, Ford, Mizoguchi and Ozu – have perished. It was only through the foresight of connoisseurs like Henri Langlois, founder of the first film archive, the Cinémathèque, that we have much of what survives. Not only did Langlois collect film, but he disseminated it through screenings, sparking interest, debate and further work in the area, trying to open up what canons and best-of polls attempt to limit, the world of film. How many people today have seen, or even heard of, once-celebrated films like La Maternelle (Jean-Benoit Lévy and Marie Epstein, 1933), The Eternal Mask (Werner Hochbaum, 1935) or The Dybbuk (Michal Waszynsky, 1937)? It was ‘forgotten’ films that were the special pleasure of Langlois and all cinephiles, adding a thrill of detective work or treasure hunt to the Idealism I mentioned earlier – film lovers (as opposed to film buffs, only interested in tedious facts and anecdotes) spend more time dreaming about unseen films than watching those available to them. This is, perhaps, and perhaps immaturely, a strategy of evading control. The very eccentricity that led Langlois to collect abandoned celluloid, however, is blamed for the erratic archiving that saw material lost due to the poor storage of flammable stock. It is in the areas of archiving and film history that the battle against the lost film has been waged, and much brave and exhaustive work has been done on recreating vanished art, whether through the restoration of existing films, or reconstructions of missing footage with reference to screenplays, stills and studio records. But even this task has something Sisyphean about it, as witness the fraught history of Metropolis (Fritz Lang, 1927) and its many versions, or even a recent George Eastman House restoration of The Lost World (Harry O. Hoyt & Willis O’Brien,1925). which excised politically incorrect material. Recreating the silent era is further impeded by controversies over projection speeds and the loss of most contemporary music scores. It is no surprise that the increase in such activity, from the late 1950s onwards, should coincide with the growth of (stop yawning) post-modernism. Without labouring the point, the lost or fragmented film is in many ways the ideal post-modern artefact, with its subversion of authorial intention and its decentring of narrative, formal, thematic and stylistic totality: the lost film, in a sense, embodies its own de(con)struction. Closure is denied, loose ends float free, the unities are blasted open. Instead of one unified production of a Great Artist, the lost/fragmented work offers a (great post-modern jargon) palimpsest of variant texts – what we have left, what was intended, what has been worked over by other hands etc. But this is labouring the point, and nothing new – the Romantics prized the fragment as a more genuine revelation of feeling then the finished work (was it premonition that saw Welles quoting Coleridge’s unfinished ‘Kubla Khan’ in Citizen Kane?), and didn’t see it as incompatible with notions of the heroic outsider artist. It should not be forgotten either that most of the classical culture on which Western Civilisation models itself is today either lost or ruined – perhaps it was such a fate that prompted Gilbert Adair to declare the lost film “ennobling.” Fragments, in any case, can often stimulate artistic design: many famous directors, such as Marker, Godard and Fassbinder, envision(ed) their films as works-in-progress; collagists such as Makavejev use the interplay and disjunction of film fragments to create new, but not necessarily ‘complete’ works of art. A work like Welles’ Othello (1952), shot over two years in intervals in between the acting jobs taken to fund the project, is fragmentary and complete at the same time. Of course, film was not the only 20th century art form prone to loss and decay in a supposed age of mechanical reproduction. One example of the unfinished artwork I have not yet mentioned is that left incomplete after an author’s death, such as Proust’s Le temps retrouvé (1927) or Berg’s opera Lulu (1935). Perhaps the most moving 20th century example of death and the unfinished artwork confounding authorial intention is Hergé’s Tintin et l’Alph-Art (1986) which leaves its perennially youthful Houdini hero facing the firing squad, unrescued and unrescuable. Famous films by directors unfinished at death are rare, such is the commercial nature of cinema, but there are many examples of directors being replaced for various reasons (health, personality clashes, mismanagement), leading to one vision being buried under that of another (e.g. Preminger replacing Mamoulian on Laura (1944); Jewison taking over from Peckinpah on The Cincinnati Kid (1965). The very idea of the lost film, and its comparative rarity, suggests that most movies are in some way ‘complete.’ But, again, and forgetting Ferrand’s complaint in La nuit americaine (François Truffaut, 1973) that a director begins a film hoping to make a masterpiece, but is soon solely concerned with getting it finished – implying the finished product is rarely the lost, ‘complete’ film in the director’s head – the industrial and international nature of film makes completeness impossible, especially in an age of multiple media and the often vast distances between supply and demand. From the moment a director calls a final wrap, a film can be declared ‘lost,’ subject to the whims of preview audiences, shrunk onto video and DVD (not just when panned and scanned), fragmented by ads on TV (which is why George Stevens took a US network to court over a screening of A Place in the Sun ). Films with local appeal (e.g. German comedy) or steeped in local reference (e.g. Paradjanov’s The Colour of Pomegranates ) can be lost in translation. Films are dubbed and often re-edited for different markets. Censorship can have a direct and indirect effect on the completeness of films – direct in the sense of cutting offending material, indirect in that filmmakers working in climates of censorship (whether in totalitarian regimes such as the USSR or Iran, or in Hays Code Hollywood) have to rethink their artistic strategies, are forced into an oblique aesthetic and, perhaps, in the process, lose something of their original sensibility. For many years, A Clockwork Orange (1971) was a famous ‘lost’ film in Britain and Ireland when director Stanley Kubrick withdrew it over fears of copycat violence, even though it was freely available in other countries. Director’s cuts and the increasing bad faith of filmmakers in tampering with works long after their release (e.g. Apocalypse Now Redux [Francis Ford Coppola, 2001]) means that today, few films can lay claim to being stably finished. And I haven’t even mentioned remakes or ancillary works created by advertising. But perhaps the greatest joy of the lost film is that in many cases it is only ‘lost,’ gathering dust in lost property offices in some of the most obscure parts of the world. A recent restoration of Buster Keaton’s complete short films by the French company Lobster Films unearthed gems long presumed destroyed by rooting around archives in Russia, for example, and the Netherlands. This kind of news only fuels ever more tantalising cinedreams. There are six basic types of lost film: 1. The irretrievably lost film, of which no print exists, and whose survival depends on a handful of stills, listings in studio records etc. King Vidor’s Bardelys the Magnificent (1926) is a famous example among the reservoir of silent films now vanished. 2. A film that exists only in fragments of (often highly degraded) footage, such as Fatty Arbuckle’s Moonshine (1918). 3. The film that never was: productions shut down for various reasons – financial difficulties, cast illnesses/death etc. These films live on in the form of rushes (which often illuminate the artistic practice of film-making more successfully than any finished work could ever do) and hearsay, and can make excellent documentaries dramatising the conflict between art, commerce and fate. Famous example: Sternberg’s I, Claudius (1937). 4. The long cherished, much discussed project that, for whatever reason, never got off the ground, surviving in biographies or unfilmed screenplays – Kubrick’s Napoleon film; Losey and Pinter’s version of Proust. Nabokov’s fantastic, unfilmable screenplay for Kubrick’s Lolita (1962) might fit into this category. 5. The film that was re-edited after previews, but of which the edited footage no longer exists, precluding the relatively modern phenomenon of the director’s cut, e.g. Lang’s Metropolis (1927). 6. The most famous/celebrated/reviled example of the lost film: the movie that is taken out of its genius director’s hands and re-edited by commercially-minded philistines. The patron saint of such ruins is the man I now intend to discuss, Erich von Stroheim. * * * It is fitting that the unfinished Queen Kelly should have achieved an ironic afterlife as a fragment. It is the footage demented silent starlet Norma Desmond screens for her reluctant protégé Joe Gillis in Sunset Blvd. (Billy Wilder, 1950). Desmond was played by Kelly‘s star Gloria Swanson who in the earlier film was nearing the end of her heyday as Hollywood’s most successful actress; the role of her projectionist, manservant, ex-husband and ex-director is silently suffered by Erich Von Stroheim, whose last major production as a director Kelly was. This casting and its echoes are all part of the in-joke torture chamber that is Wilder’s film, even if Kelly is funnier, more scabrous and psychologically probing than anything in that Gothic cartoon. The pay-off of this particular in-joke is that producer Gloria Swanson sacked director Erich Von Stroheim on Kelly, supposedly for his legendary extravagances. I shall suggest later that perhaps Swanson’s piqued vanity may have had something to do with it. Because Queen Kelly was a vanity project, a prestige product for an actress who’d recently become an independent producer; it was financed by her lover Joseph Kennedy, bootlegger, father of JFK, and the man who saved the company that would become RKO, even if he didn’t know the first thing about movies. Kelly was to have been an actress’s tour-de-force, in which Swanson would have to master all styles from comedy to tragedy, all emotions from love to degradation. Kennedy had been warned off Stroheim by almost every Hollywood money man who’d had dealings with the Austrian – with his monomaniacal obsession with detail, his relentless accumulation of footage and his taste for ‘sordid’ subject matter. But Kennedy was canny and insecure enough to intuit that immortality was more likely to be bestowed by association with a wayward genius than mere money-making ephemera. So, Stroheim’s treatment, unpromisingly entitled The Swamp, was given the go-ahead, with two producers hired to make sure the Von stayed reasonable. To Swanson’s recorded dismay, they failed to protest as endless and seemingly undifferentiable retakes began eating up the budget, already high after the erection of elaborate sets recreating a turn-of-the-century Ruritanian kingdom (she also feared that Stroheim’s increasingly bold sexual frankness might not escape censorship). It didn’t help that Kennedy had suggested the film, which went into production in 1928, be shot silent, despite the increasing hoopla over sound pictures. At the rate Stroheim was shooting, the film threatened to be obsolete before it was even released. Stroheim, at the mercy of his domineering star/producer just as his weak-willed hero must submit to a power-crazed, luxury-bloated Queen, was summarily dismissed. A journeyman director, Richard Boleslavsky, was brought in to take over, but even Swanson had to admit that this hack couldn’t begin to approach the mercurial majesty of Stroheim’s rushes, and the plug was finally pulled in 1929. Swanson, by this stage deserted by her gangster consort, released her version in 1931, abridging further the available footage so that her heroine could chastely commit suicide, thus avoiding the gleefully sordid humiliations of the brothel section. Stroheim released his own version in the 1960s, and a more thorough reconstruction, based on the original screenplay, and featuring an ‘epilogue’ of stills and intertitles, was put together in 1985, which is the one available on the recently released American DVD. Originally intended to last 30 reels or five hours, roughly a third of Kelly was shot. In spite of its abruptly curtailed state, however, Kelly is in some ways a more satisfying viewing experience than more acclaimed Stroheim masterpieces such as Foolish Wives (1921) or Greed (1925). Those films had been completed by Stroheim, were taken away and insensitively re-edited, destroying their meticulous visual and narrative design. The first section of Kelly, at least, is mostly intact, and is as close to ‘pure,’ mature Stroheim as we’re likely to get. The plot of Queen Kelly combines the Faberge spectacle of costume dramas like Foolish Wives or The Wedding March (1928) with the inexorable squalor of a Greed. The three sections as we have them offer the experience of watching three different kinds of lost film – part one is virtually complete, suddenly abbreviated at the end; part two survives as a long fragment; part three is synopsised by stills and titles. Part one is set in Kronberg, capital of an ancient Middle European kingdom, ruled by the ‘mad’ Regina V (Seena Owen), said to share the blood mania of her decadent line. This is a woman who has refused her appetite nothing in the way of sex, power, luxury or spectacle, and the surfeit has gone to her head, accounting for the dazed look with which she sleepwalks through the film. We first see her awakened on her large, lavish bed smothered with pillows, surrounded by the detritus of her debauchery (Boccaccio’s Decameron, Veronal) her modesty protected by a cat that is both a comically obvious figure for her preening sexuality and a forewarning of future events when, in richly pouffed robe, she prowls the castle in search of her fiancé. This is her cousin Prince Wolfram (Walter Byron), known as “Wild Wolfie.” He too is introduced as captive to a decadent sexuality as he drunkenly and suggestively rides a tightly reined horse for the stimulation of a carriageful of ladies. But whereas the Queen’s sexuality is passive, indoors, sapped of energy, the result of appetite too freely indulged, Wolfram’s is active, outdoors, natural, the result of thwarted energy diverted. The Queen, mocked by the revellers and affronted by Wolfram’s apparent freedom, brings the wedding date forward to the following day, and punishes her relative to early morning manoeuvres with the horse guards. This rustic exercise coincides with the daily walk of the local convent orphans, and as the admiring Prince passes the girls he is particularly struck by Patricia Kelly (Swanson), the spirited daughter of a wandering Irish artist. When her panties fall down as she curtseys, the Prince is amused, and she hurls them at him in a rage, too naïve, perhaps, to understand how welcome such an insult might be; and, indeed, the Prince, bayonet erect, holds them near his nose. As the nuns mutter in outrage, and the soldiers begins stalking the girls, the Prince and Patricia build up a kind of bantering relationship, purely through facial expression, resulting in both making a wish on sods of newly mown hay. Similar chastisements await both when they arrive ‘home’: the Prince is informed of his impending wedding; Patricia is grounded for a month, and refused that evening’s supper. She prays to the Virgin and Child to send her the Prince, and, sure enough, depressed at his doom, Wolfram decides to have one last bachelor fling. With the help of a friend he abducts Patricia by faking a fire in the convent, wines and dines her, finds, perhaps, that his feelings are more profound then he’d realised, but still plans to deflower her (fortunately, the 30-year-old Swanson looks her age by this point), only to be interrupted by the frothing Queen who horsewhips them both, driving Patricia to attempt suicide. She is rescued by a guard, and, on return to the nunnery, finds a telegram calling her back to a dying aunt in German East Africa. Wolfram is meanwhile put in solitary for his refusal to renounce the schoolgirl. The perversions of the first section are stripped of their high-class trimmings in an African sequence reeking of death, disease, deformity and depravity. On her deathbed, Patricia’s aunt, madam of a squalid brothel, orders her niece to marry her crippled right-hand man, Jan (Tully Marshall), who has been waiting for the crone to croak so he can take over. In a deliciously grotesque sequence, trumping the famous wedding/funeral set-piece in Greed, extreme unction is succeeded by a wedding performed over the deathbed climaxing with the aunt’s demise, Patricia’s sinking acknowledgment of her debasement and Jan’s hobbled rising to power. The epilogue recounts the substance of the remaining, unfilmed two-thirds – Kelly becomes the new madam, the Prince released from prison goes searching for her, Jan is killed in a brawl, the Queen is assassinated, the Prince takes Kelly home to rule in Kronberg. It is tempting to read the broken state of Kelly as a response to its own themes – like Touch of Evil, another famously tampered-with masterpiece, its narrative proceeds from a coitus interruptus, when Kelly and Wolfram in the Prince’s bed are intruded on by the Queen. Where Welles’ film heals the initial rupture, however, the unfinished Kelly cannot, thus frustrating the close links noted by Foucault between narrative and sexual closure. The narrative as filmed begins to break down at the moment of greatest emotional trauma for its characters, as if attempting to accommodate this disintegration in the very form of the film itself. The maimed African sequence is peopled by characters dying or wasting away, their bodies (and souls) broken beyond repair. But this is to gloss over the irreparable damage done Kelly, and to overlook the coherence of Stroheim’s design as evidenced in the first section. Firstly, there are a number of image or motif systems whose ultimate interaction we can only dimly guess at, but whose often unexpected evolutions provide much of the first section’s momentum. The most blatant is the motif of fire, especially as channelled by candle, which as both dripping erectile object and a more spiritual means of illumination, conflates the twin themes of sexual and religious passion, initiated when Patricia, inspired by a naked Christ (the ‘King’), lights a candle in the convent chapel praying to see the Prince once more. Next, masses of candelabra illuminate the announcement of the Prince’s unwanted marriage to the Queen. The lighting of a cigar by candlelight leads Wolfram’s friend to the hay which spurs the Prince to his plan, and, fittingly driven by amour fou, Stroheim appropriates the Surrealist method of making the symbolic concrete: it is by fire that Wolfram diverts the nuns and kidnaps Patricia. Increasingly suggestive candelabra frame the supper scene as the Prince contrives to seduce Kelly, with the fire raging behind her encouraging her to strip, eventually leading to the bedroom, the Queen’s wrath and the major narrative plot point. This motif also contributes to an ironic contrast between light and darkness in the film, elaborated in the chessboard floor of the Queen’s palace, and culminating in the racial tension of the wedding scene, as the depraved colonialist is married to an Irish convent girl by an African priest, with black and white prostitutes-as-bridesmaids sneering at the sides. With this sequence curtailed as it is, we cannot know how this fraught subject matter might have developed. Other image- and thematic systems which interact in counterpoint include animals, costumes (especially dressing, undressing and cross-dressing), fairy tales, literal rising and falling, and mirrors, all of which are used to explore a contrast between the rigid roles imposed by society, and the freedom of multiple identities available through sex. This Utopian reading, however, is qualified by Stroheim’s use of mirroring as a structuring device in Kelly. Patricia may eventually yield to the Prince, but this is the result of an elaborate trap that begins with abduction and utilises all the resources of ‘culture’ – the dazzling décor (including pointedly erotic paintings) in Wolfram’s rooms that so overwhelms Patricia she has to open the door for air and enter a grove that is cunningly contrived to suggest the ‘natural’ in contrast to the artifice within, thus furthering Wolfram’s aims. To get to this point, just as he and his soldiers stalked the convent girls, so the Prince relentlessly pressures Patricia, plying her with alcohol, extracting her sexual fantasies, hiding her outdoor clothes, dogging her around the room. This matches Wolfram’s depressed decadence earlier, and if the Prince does become genuinely touched by feeling, we shouldn’t let the poncy costumes and elaborate furnishings blind us to the fact that Wolfram’s predatory antics are mirrored by the hideously lecherous, lizard-like Jan in the African sequences (both are heavy drinkers; both are ambitious men bridling in female-centred establishments; both are aided in forcing Patricia to submit by the pressured editing of the director). Stroheim, of course, knew the snob in all of us would be blinded by finery, which makes the African sequence such a slap in the face, one that certainly bruised Swanson , who didn’t seem to mind being debauched in costume, but drew the line at gibbering cripples. Kelly herself makes the connection between the two worlds when, cracking under the emotional pressure of the ghastly wedding ceremony, she replaces in hallucination the priest with the supplicating Wolfram, again opening up extraordinary thematic possibilities never to be explored. Symmetry creates the most unexpected connections throughout the film, from the pair of forced marriages, to the links through ceremony and uniform between the convent and the army, to the many reflections and patterned contrasts between the Queen and Queen Kelly and their alternating rise-and-fall/fall-and-rise narratives. These two sequences and their internal architecture give some indication of how a ‘complete’ Queen Kelly might have evolved, although Stroheim as a director always enjoyed mixing an illusion of narrative inevitability with play on an audience’s expectations. But Kelly as we have it is not Stroheim’s film, but Gloria Swanson’s. Her decision to stop production may have been the result of socio-economic and historical forces, but it was also a reading of the film that supplanted the coherence and generosity of Stroheim’s vision. Even if we reject Swanson’s released version – surprisingly conventional, Victorian, melodramatic and anti-feminist, with the suffering heroine killed off rather than struggling (no matter how ironically Stroheim may have intended such a narrative) to achieve status and economic independence – the butchered remainder in no way matches the totality of Stroheim’s original conception (obviously). Because his works were continually interfered with, Stroheim was forced against his temperament to become a modern (his surviving films are decentred, disjunctive, fragmentary, incomplete, open), when he probably yearned to be the cinematic equivalent of a great 19th century novelist (omniscient, totalising, exhaustive, in complete command of his medium and material, closed). Kelly not only killed off Stroheim’s career as an auteur, but confirmed that Stroheim could never really be considered an auteur in the purest sense. The original politique des auteurs praised those directors who adapted their artistry to the studio system, achieving from within the control Stroheim craved. The great films of Hitchcock, Hawks, Lang, Sirk and Mann exist as their makers intended, whole and free; Stroheim’s only in the ideal, imagined sense. Already, as those others grow in reputation, Stroheim begins to fade, no longer ranking in ‘Best Of’ lists. The films left us today are not signed by him, but by the corporate professionals who fancied that their vision, because aimed at an imagined mass audience, was broader. Bibliography Gilbert Adair, Flickers: An Illustrated Celebration of 100 Years of Cinema, Faber and Faber, London, 1995. Seth Bookey, “Queen Kelly (1929)” (www.reviews.com/Reviews/229/22964) Serge Bromberg, “Les restaurations du Monde perdu (1925)”, cinémathèque, Autumn 2001, pp. 25-35. Kevin Brownlow, “Silent Films: What Was the Right Speed?”, Sight and Sound, Summer 1980, pp. 164-167 (www.cinemaweb.com/silentfilm/bookshelf/18_kb_2.htm) Ivan Butler, Silent Magic: Rediscovering the silent film era, Columbus Books, London, 1987. Sandrine Fillipetti, “La restauration des ‘courts métrages’ de Buster Keaton”, cinémathèque, Automne 2001, pp. 5-24. Gavin Lambert, “Stroheim Revisited: The Missing Third in American Cinema”, Sight and Sound, April/June 1953 (www.bfi.org.uk/showing/nft/featurearchive/stroheim/lambert.html). Jack Lodge, John Russell Taylor, Adrian Turner, Douglas Jarvis and David Castell, Hollywood: 50 Great Years, Prion, London, 1989. Peter Noble, “Stroheim – His Work and Influence”, Sight and Sound, Winter 1947/8 (www.bfi.org.uk/showing/nft/featurearchive/stroheim/noble.html). Harvey O’Brien, “Queen Kelly (1929)” (http://indigo.ie/~obrienh/qk.htm) Liam O’Leary, The silent cinema, Dutton Vista, London, 1965. Jonathan Rosenbaum, BFI Classics: Greed, British Film Institute, London, 1993. Christopher Silvester (ed.), The Penguin Book Of Hollywood, Viking, London, 1998. Gloria Swanson, Swanson on Swanson, Michael Joseph, London, 1981. David Thomson, Rosebud: The Story Of Orson Welles, Little, Brown, London, 1996. Parker Tyler, Classics of the Foreign Film: A Pictorial Treasury, The Citadel Press, New York, 1962.