Dismembered Bodies, Dismembered Frames

Peter Greenaway’s The Pillow Book (1996) is a film about bodies. More precisely, it is a film about dismembered bodies. Part of the climax of its gruesome revenge story involves the flaying of the corpse of Jerome (Ewan McGregor), erstwhile lover of both the film’s protagonist, Nagiko (Vivian Wu) and antagonist, the corrupt publisher (Yoshi Oida). Nagiko, a Japanese born model, chooses lovers—including Jerome—who are willing to write on her body, and whose skin she can use as pages for her elaborate series of thirteen cryptic calligraphic books. When Jerome dies in a faked suicide that goes wrong, Nagiko writes her final text on his corpse. The publisher steals the corpse and skins it in order to preserve the text, but Nagiko wreaks her revenge by having the publisher killed and retrieving the book. And all of the dismembered Jerome that is not skin ultimately, and with an outrageous lack of ceremony, ends up in a rubbish bin.

This shocking conclusion plays out a recurring motif in Greenaway’s films, namely physical dismemberment as literalisation of central narrative metaphors: Neville’s blinding in The Draughtsman’s Contract (1982) robs him of both sight and insight; Alba’s second amputation in A Zed and Two Noughts (1985) seeks to restore both physical symmetry and emotional equilibrium; Albert’s carving and consumption of Michael’s roasted corpse in The Cook, the Thief, His Wife and Her Lover (1989) is both cannibalism and a performance of gangster greed. These violent disruptions of bodily integrity stand as the climax of key thematic trajectories. Throughout The Pillow Book, the film’s intimate exploration of the characters’ physical forms—as aesthetic objects and narrative spaces—depicts ways in which both erotic desire and language carve up bodies; Roland Barthes argues that ‘language [including film] is, as it were, that which divides reality (for instance the continuous spectrum of the colours is verbally reduced to a series of discontinuous terms).’ (1967: 64) Long before skin is literally detached, the film returns obsessively to synecdochic close-ups in which skin is both isolated from and called upon to represent the whole body. Nagiko’s skin-deep desires for sex and revenge drive the narrative, and every stroke of her pen is a cut into the skin on which it writes.

This article tethers this interest in dismemberment, as both a narrative event and a narrative strategy, to film form. Just as it cuts up its narrative bodies, The Pillow Book hacks its overall frame into smaller elements, arranging written text, painting, still photography, and moving celluloid, digital, and video images into an illegible pile-up of intermedial parts. As Nagiko reflects on sex with an Englishman, the image track consists of three discrete visual planes: at the base, the moving image of the man’s naked body has superimposed over it a layer of text (both painted Japanese calligraphic letters and Arabic numerals), which, in turn, is part-obscured by the opaque film image of upturned bowls, describing the man’s belly in an imperfect metaphor (Figure 1). The combination of layered images in this densely saturated frame is further complicated by music and a voiceover that quotes from Sei Shōnagon’s Pillow Book, the inspiration for Greenaway’s film.

Figure 1

The dismembered intermedial frame in The Pillow Book (1996).

A Japanese writer and court lady to the Empress Teishi, Sei Shōnagon is most celebrated for The Pillow Book, which consists of poetry, instruction in rhetoric, gossip, and lists. It offers instruction in a number of courtly pursuits, including letter writing, extracts from which are quoted in Greenaway’s film. The book records the quotidian and the fleeting, and has been subject to significant editorial interventions through the centuries; it was only set in print in the seventeenth century and exists in many different versions, each of which orders, comments on, and excises materials differently (see Keene, 1993). The exact details of Sei Shōnagon’s birth and death, and even her given name, remain uncertain. Greenaway’s earlier adaptation, Prospero’s Books (1991), takes its inspiration from The Tempest by William Shakespeare, another writer whose life remains shrouded in some mystery, but Sei Shōnagon is an even more enigmatic figure. This enigma is reflected in the complex form of adaptation used by Greenaway, an unpredictable response to the source text that further cuts up and confuses Sei Shōnagon’s original work. As Nagiko considers her English lover, the source text, quoted in combination with a frame that groans under the weight of so many images, does not offer stability but another set of complicating fragments.

This dense piling up of dismembered signifiers—with an emphasis on multiple frames, approximate metaphors, untranslated text from multiple languages, and enigmatic literary allusion—results in a form that undercuts narrative or thematic coherence. Nevertheless, dismemberment as both metaphor and formal strategy has been overlooked by most criticism on The Pillow Book and on Greenaway’s films more generally. Santiago Fouz-Hernández does not consider the extent of the formal complexities in his reading of the film as a study in ‘gender equality that celebrates the emancipation of women’ by ‘recuperat[ing] the original, underlying female voice (Shōnagon’s) and culminat[ing] in a matriarchal closure.’ (2005: 141) This insistence on thematic and narrative closure runs through a body of criticism that has focused on how the film’s form—and in particular its combination of images and sounds through the use of digital technologies—has, in Brian Hoyle’s words, ‘create[d] something whole and new.’ (2016: 327) This strand of analysis began in Claire Tovey’s review of The Pillow Book, which defines what it describes as the film’s ‘utopian’ form, one that ‘can bring together’ past, present, and future, and word and image. (1996: 58) This article argues that, far from bringing things together in a utopian whole, The Pillow Book formally dismembers its frame into parts that remain imperfectly reconciled. Utopian synthesis is not, I argue, a quality of the film but rather the conclusion of criticism which, like Hoyle’s, argues that with ‘repeat viewings’ the film will ‘divulge all of [its] secrets.’ (2016: 319) I suggest, instead, that The Pillow Book’s form exceeds legibility and, in so doing, frustrates the classificatory systems (including those of critics) which may seek to marshal its dismembered body into meaning.

A discussion of Greenaway’s formal dismemberment features more prominently in work by Bridget Elliott and Anthony Purdy, who relate the ‘daunting’ number of ‘visual quotations’ to the director’s interest in bodies as objects that ‘are always already multiply inscribed by the systems and narratives and images that construct them.’ (1992: 201 and 204) However, while they recognise Greenaway’s willingness to ‘manipulate and even dismember the human (or animal) body’, they ultimately define the form according to regulated ordering systems that use ‘the body as a privileged site for (richly comic) articulation of the intelligible and the sensory.’ (1992: 204 and 205) A similar sleight of hand, which registers the startling dislocations of Greenaway’s form but also finds ways of synthesising its fragments into coherent wholes, is at work in Paula Willoquet-Maricondi’s work on Greenaway’s sceptical critique of written language. Willoquet-Maricondi presents writing as a kind of dismemberment of the world, one that alienates the human subject from nature by establishing an objectifying distance. This engagement with Greenaway’s cinema incorporates the Oedipus legend, gender politics, and poststructuralist theory in order to complicate the relationship between text and body in the films, but nevertheless only reluctantly agrees ‘that it could be argued that Greenaway’s text, The Pillow Book, manages to retain a certain degree of ambiguity as to the ultimate meaning’. (1999) Willoquet-Maricondi remains invested in character analysis as a means of justifying the formal complications that represent Greenaway’s ‘ongoing efforts to find a suitable union of image and text in his elaboration of a non-narrative cinema.’ (1999)

I argue, instead, that The Pillow Book’s dismembered form marks the articulation of the not-quite intelligible and the sensorily overwhelming, and that image, text, and body—and, with them, meaning—are fragmented and not united by the film. This article aims to show that when criticism attends to the film’s specific formal achievements, rather than privileging the discussion of its content, a complex critique of systems of knowledge becomes visible.

Eisenstein in Greenaway

The Pillow Book’s complex form—in which myriad elements are combined—suggests an affinity to the theories and works of Sergei Eisenstein, and it is perhaps unsurprising that Greenaway’s most recent film, Eisenstein in Guanajuato (2015), centres on the Soviet filmmaker. Eisenstein, like Greenaway, explores the productive correlation between corporeal dismemberment and film form, comparing cinematography to ‘[h]ewing out a piece of actuality with the ax of the lens’ before ‘montage disintegration.’ (1949: 41 and 34) In Montage 1937, Eisenstein explicitly likens edited film form to the violently cut up and reformed human body. Yuri Tsivian summarises the position:

Eisenstein had observed that, by its very technical nature, film-making involves two operations that mirror each other: at the stage of production the director fragments ‘reality’ into shots only to reassemble them, in a different manner, on the editing table—the break-and-make process… [T]he underlying site of a true tragedy, regardless of its setting, is the human body…; while its ultimate action—the act of dismemberment and subsequent welding—is montage… From ancient times, says Montage 1937, art has known the secret of the ‘Osiris method’ (here Eisenstein evokes the Egyptian god deemed to relive, seasonally, dismemberment and resurrection)—the precarious equilibrium (or, better, the dynamic tension) between unity and disunity, the centrifugal and centripetal. (2002: 49–50)

Conventional continuity editing restores integrity to the body with such apparent seamlessness that the dismemberments which are an inevitable part of filming are largely masked. By contrast, The Pillow Book, through a series of ostentatious formal achievements, foregrounds its own playful-earnest assault on the linear structures of such narrative film convention; the body remains in pieces and the film’s dislocated images and sounds become like so many severed limbs. This re-memberment results in a complex frame that is reminiscent of the grotesque out-of-orderedness of a cadavre exquis. A sense of integrity is abandoned in favour of an dismembered form which makes suture (a term happily borrowed from surgical processes on organic bodies) impossible.

The Pillow Book’s process of disorientating re-memberment is foregrounded through its making-visible of the synthetic properties of film, again resonating with Eisenstein’s theories, this time of film’s synthesis of other art forms into a total artwork. In 1995, a year before The Pillow Book was released, Greenaway discussed this question in the catalogue for his city-sized installation, The Stairs: Munich Projection. Despite ostensibly being a celebration of cinema’s centenary, Greenaway’s introduction to the exhibition reads more like a post-mortem, saying that for the medium to ‘stay vigorously alive… has now become improbable.’ (1995: 15) This air of ‘apparent morbidity’ is relieved by a sense that this statement is a kind of autopsy-as-resuscitation, a dissection to allow for reformation; it celebrates the emerging possibilities of a ‘mega-cinema’, in which intermedial forms might transform film. (1995: 11) Greenaway writes that:

The cinema, they say, is the medium for all seasons. Perhaps it is not a single medium at all, but a plethora of possible ways to communicate hoping for synthesis. It is writing illustrated, theatre recorded, the talking novel, painting on the move, image serviced to music. (1995: 13)

Given that theories of film as a synthetic medium have circulated since the 1920s, this may seem like an unremarkable claim. Eisenstein describes film as ‘that genuine and ultimate synthesis of all artistic manifestations’, beginning the theoretical interest in intermediality, which has developed in line with technological advance. (1949: 181) However, instead of constructing frames from mutually-reinforcing synchronised images, sounds, and texts, and arranging these frames linearly in synthesised or sutured sequences, The Pillow Book consistently—though not systematically—piles up non-synchronised frames in unstable towers. The overall frame is regularly broken up by multiple insert-frames layered on top of or beside each other, accompanied by an eclectic sound score consisting of Japanese folk music, European pop music, diegetic sound, non-diegetic noise, and voice-over. These piles of frames make visible how easily the signifying chains that make up a film—so often simplified by the linearity of continuity editing—can be rattled.

For Eisenstein, these signifying chains are predominantly conceived of in diachronic terms, extending linearly through time. While he does discuss the possibilities and effectiveness of simultaneous or vertical montage, video and digital technologies, which have emerged over the last thirty years, have permitted Greenaway, ‘since the release of A TV Dante (1990) and Prospero’s Books (1991)’ (Hoyle 2016: 314), to take these ideas further than Eisenstein could. Multiple insert-frames—consisting of film, video, hdTV, and digital imagery in combination—are gathered together in a densely saturated frame that exceeds the neat dovetailing of narrative logic and a coherently depicted film world. These frames do not respect chronology, upsetting the primacy of the one-thing-at-a-time structure of cause-and-effect narrative film as their constituent audio-visual elements refuse to unify. The Pillow Book’s ostentatious combination of writing, painting, photography, and moving image within a frame dismembers the diegetic world (including moving images of Nagiko’s childhood, moving images of her marriage, moving images of her adult life in Hong Kong, colour images of Sei Shōnagon’s 10^th century court life, still photographs and postcards, static text in different languages, all presented in a dizzying array of aspect ratios) and openly represents it.

In classic film syntax, these frames would be obliged to play out sequentially, through time and across navigable space; deviations from temporal linearity would be justified narratively, through the use of flashbacks, for example. In The Pillow Book, they are snatched out of this temporal, narrative relationship and reimagined in a spatial relationship. This marks a technologically-enabled shift from succession to simultaneity. By making the usually unnoticed transitions and combinations visible, The Pillow Book shows ‘how films are “put together”’, as James Tweedie expresses it. (2000: 111) What is being made visible here is not simply the fact that films have constituent parts, but that those parts are actively arranged in order to create meaning; there is always-already an invisible agency which has ‘“put together”’ the text. The Pillow Book’s editing is openly performative, insisting that its own structures be noticed as unnatural in a statement on the performativity at the root of all film form as it reproduces a series of indefinitely repeated and ritualised structures that disavow the processes through which apparently natural structures are constituted and give the impression of significance.

This performativity makes visible the hierarchy implicit in all film re-membering, in which frames that are not selected are typically left invisible. The Pillow Book, unusually, shows multiple frames, exposing how certain images dominate and eclipse others. The film rarely uses superimposition (laying one semi-transparent frame over another in such a way that both remain evident) and instead favours a kind of superposition, in which opaque frames are layered over each other so that the upper frame obscures the lower, with some frames all but buried behind others; these competing frames jostle for the privilege that comes with scale and position, as legibility is dependent on being either larger or layered nearer the front of the overall frame. This competition exposes the typically concealed violence of the privilege and suppression at work in the world- and meaning-building operations of film form. Yvonne Spielmann argues that:

The coherent image makes invisible the techniques of merging and layering. In opposition, Greenaway’s incoherent image is categorized as intermedial essentially based on the differences that can be recognized. (2001: 59)

The recognizable differences allow The Pillow Book’s frame to reveal the ideological sleight of hand at work in continuity editing. The competing frames cannot be straightforwardly stitched together as their contiguity exposes an aesthetics of exclusion. Conventional formal integrity is no more to be depended on in The Pillow Book than bodily integrity is in a charnel house.

This use of superposition is an assault on the conventions of both filmic space and time, and the underpinning myth of cause-and-effect plotting. The insert-frames distract from or cover up images of narrative importance by upsetting the usual configuration of the space of the frame, and this new spatial simultaneity is dislocated from any sense of temporal simultaneity. Temporally disordered frames coincide in an overall-frame in which narrative is consigned to secondary importance. In the sequence in which Nagiko is depicted being written on and slept with by a writer while his wife sings in the next room, the linear consecutiveness of writing and sex is lost and the two events occur together in the frame, while his wife, in a small insert, hovers in the top right corner (Figure 2). The horizontal, syntagmatic axis of sequential combination is deemphasised, and the vertical, paradigmatic axis of selection and association is strangely manifested with the making-present of should-be-absent signifiers (in the literalisation of the pen/penis metaphor and the revealing of the meanwhile-elsewhere wife). Causality (Nagiko sleeping with the man because he will write on her, and the wife singing despite her husband’s infidelity) is compromised in a disordered, non-sequential narrative. Sex dominates as the largest single image, but writing—layered above it—obscures it, obliterating precisely the central image of penetration. The back-layered action is behind a top layer, while the top-layered action—an act which narratively culminates in the sexual activity of the back layer—is dwarfed.

Figure 2

Disturbed linearity in The Pillow Book (1996).

In this dense frame, the parts both cohere and compete, both centripetally hold together and centrifugally pull apart, performing both the dismembering and re-membering gestures which have organised them. The frame collapses the dichotomy established by André Bazin between ‘Painting and Cinema’:

The picture frame polarises space inwards. On the contrary, what the screen shows us seems to be part of something prolonged indefinitely into the universe. The frame is centripetal, the screen centrifugal. (1967: 166)

Through an intimation of an off-screen world (beyond the ‘piece of masking that shows only a portion of reality’), the film screen pushes outwards. (André Bazin, 1967: 166) This unidirectional energy, Bazin argues, contributes to film’s intimate connection with reality, a connection from which the images’ value may be derived. In The Pillow Book, the inserted frames—often stylised, contrived, and isolated from the masking of the screen edge—attain a kind of centripetal force, severed from any reality beyond. The larger, back layers, meanwhile, often retain the sense of centrifugal force identified by Bazin, pulling out and apart from the hermetically sealed inserts that eclipse parts of them. This saturated, centripetal/fugal frame becomes a stuttering complex of inconclusive utterances—‘that hybridization, mestizaje, dialectic, whatever you want to name the process that makes language’—enounced as a torrent of sounds and images. (Lawrence Chua and Peter Greenaway, 1997: 28) In this torrent, borderlines open up between constitutive parts, and meaning remains suspended across them, an unrealisable potential. As Paul Melia and Alan Woods write, ‘always there is what is missing, there is always the gap between two frames, the fragment between two fragments.’ (1998: 20) In The Pillow Book, these gaps are made startlingly visible as frames appear concurrently and not consecutively.

Nowhere is this more profoundly felt than in the sequence depicting Jerome’s suicide. With its shifting aspect ratios, the film has already deployed black bars at the top or sides of the frame. The eye that is used to CinemaScope films shown on television screens is accustomed to this phenomenon, and the black spaces are hardly read as ellipses. However, in the suicide sequence, the inexpressive black regions of the frame become dominant (Figure 3). The stark framing forces the images inwards, requiring a reading which must leap the literally realised interstice of the black expanse, struggling to make sense of the inserts depicting numbering. The screenplay suggests that Jerome is numbering the pills he takes, but this is far from clear in the film, and the act of taking a pill is at no point matched with the act of scoring a number. Motivation and meaning are missing, the act so risibly absurd as to be implausible; even Jerome laughs when Hoki (Yutaka Honda) first suggests the suicide plan. The histrionic flourish of downing the pills with black ink is another improbable gesture. In a suicide which is both imitative (referencing Romeo and Juliet) and original, both fake and real, blackness dismembers the screen and frames the action in indeterminacy. In the extent of this indeterminacy, The Pillow Book diverges from Eisenstein’s theories of montage.

Figure 3

Centripetal darkness in The Pillow Book (1996).

Beyond Interpretation

Eisenstein’s theories are founded on the idea that from ‘the collision of two given [oppositional] factors arises a concept.’ (1949: 37) The dialectic outlined by Eisenstein operates as the synthesis of depictable images or sounds into an undepictable concept or ideogram; images become meaningful through their fusion. While this Soviet style of montage is clearly distinguishable from the seamlessness of continuity editing, they are both committed to the discovery of meaning despite difference, the former through synthesis and the latter through suture. Greenaway articulates a different style, one which is only ever ‘hoping for synthesis.’ (1995: 13) This negative dialectic is a dismembered form but does not collapse entirely into disparate parts; it is neither meaningless nor, as Eisenstein’s theory suggests, productive of a stable concept. Carla Molinari argues that Greenaway and Eisenstein share a commitment to ‘creating meaningful sequences’ through montage, but secure meaningfulness is not so easily located in the intellectual montage of Greenaway’s frames. (2018: 336) The Pillow Book’s intermedial frame is both integrated and a number of frames at risk of disintegration.

Greenaway’s departure from Eisenstein’s commitment to the stable ideogram may be articulated through a reconception of his vertical montage form as collage. In his 1970 work on Expanded Cinema, Gene Youngblood predicted that technological advances in filmmaking would produce a form in which ‘synaesthetic syncretism replaces montage with collage.’ (1970: 85) The emerging video technology which was redefining experimental film for Youngblood allowed for a ‘multi-dimensional simulsensory network of information sources,’ fitting for the ‘nonuniform, nonlinear, nonconnected electronic atmosphere’ of the contemporary world. (1970: 77) The notion of collage has featured prominently in criticism on Greenaway’s cinema. Paul Melia sees ‘collage’ as Greenaway’s central ‘aesthetic of conjoining’, and Alan Woods argues that the use of this ‘collage space’ is a reference to ‘folk art’. (1998: 8 and 28) Peter Wollen suggests that, ‘[a]t heart, Greenaway… is a collagist, juxtaposing images drawn from some fantastic archive’. (2006: 38) Yvonne Spielmann locates in Greenaway’s work an ‘electronic collage’ or ‘cluster,’ namely ‘a type of image that is widely used in electronic media through the multiple layerings of different images or image elements, resulting in a spatial density.’ (2001: 55 and 57)

Youngblood’s prophecy resonates with The Pillow Book’s dazzling range of collage-forms, drawing on the ‘fantastic archive’ that Wollen references. From the ancient fragmented literary style of Sei Shōnagon’s writing to the simultaneity of its superpositional digital frame, The Pillow Book produces saturated collages in which discrete frame-inserts, although dialectic, cannot synthesise into a coherent meaning as they are constantly caught in the act of concealing as much as revealing; in Greenaway’s words, ‘collage never fused two or more ideas together.’ (Alan Woods, 1996: 266) This reveal/conceal dynamic is seen in the sequence when the young Nagiko observes the construction of books at the publisher’s office. Discrete images of each stage of the process are superpositioned in layered frames, just as the book’s component parts are layered together (Figure 4). This is a concentrated instance of what Spielmann describes as ‘the linear structure of film [being] disrupted by multiple layers.’ (2001: 58)

Figure 4

Discontinuous process in The Pillow Book (1996).

The sequence goes beyond the simple recognition that ‘events in reality do not move in linear fashion’ as suggested by Youngblood. (1970: 97) The competitive relationship between frames is temporally and spatially uncomfortable, as images are overtaken or eclipsed by others. This results in a confusion of emphasis in which a single event may be both back layer and top layer, the back layer both prioritised (as the largest single image) and effaced (by the images in front of it). There is simultaneity and separation, the rigidly demarcated frames keeping separate elements of the process together and apart, co-present on the screen but impossible to concentrate on in the same moment. The layered images operate as elaborate and erratic repetitions, and depth is a function of destabilising iterability, a mise-en-abyme in which, Escher-like, simultaneity makes identity (across time and in space) impossible and absurd. The delinearising of process sees elements both preserved and abandoned in a Hegelian sublation through which they both remain the same and are changed. This depiction of bookbinding holds together as a sequence, but it also reveals the conventionally invisible strings that bind books and suture films. By rearranging the pieces of a sequence into the simultaneity of collage, this sequence exposes the ‘madeness’ of meaning and the ways in which narrative tells the world into order by omitting, excising, deemphasising, and censoring.

Later in the film, the sequence is revisited, this time with a clearer parallel drawn between the dismemberment of the human body and film form. The film’s complex body-as-book metaphor, in which the skin of living lovers becomes the pages of Nagiko’s books, is ultimately deformed through the grotesque literalisation of the book fashioned from Jerome’s flayed skin. This process throws the ontology of book into question: ‘book’ seems to signify bound pages, living skin, dead and dismembered skin, and the film itself. The tour de force sequence in which Jerome’s skin is transformed into a book repeats the formal arrangement of the earlier bookbinding sequence (Figure 5). However, like all repetition, this iteration only almost duplicates its precursor. This time in lurid colour, with a key emphasis on reds, the process is synchronic, re-membering the earlier sequence from Nagiko’s childhood. Even when paused, the sequence does not lose its momentum, its corner-placed front layers pulling apart from one another, resisting a collective reading. Not only is this a grotesque sequence and the unacknowledged partner of the earlier (and then-innocent) sequence, it is also made impossible by concurrency; process becomes indeterminate if its linearity is amended to permit paradoxically simultaneous discrete positions. Jerome’s skin exists in a state outside of the possibility of narration if it is being cut, cleansed, folded, glued, stitched, marked, and etched in the same moment. Robbed of its diachronic structure, this process makes images impossibly visible in the same moment while also burying in invisibility the remembered black and white images of bookbinding from earlier in the film which are both unrecognisable in this perverse retelling and persistently foundational.

Figure 5

Re-membrance of obscene process in The Pillow Book (1996).

Kristen Daly argues that the use of combinative digital technologies ‘together… create a full experience.’ (2010: 85) In The Pillow Book, however, fullness is excess, a cacography (or visual cacophony) which has led to what Greenaway describes as critics’ ‘shrill and indignant cries of “visual indigestion”.’ (1995: 21) This is in stark contrast to Youngblood’s utopian theory of expanded cinema:

There’s no conflict in harmonic opposites. Nor is there anything that might be called linkage. There is only a space-time continuum, a mosaic simultaneity. Although composed of discrete elements it is conceived and edited as one continuous perceptual experience. A synaesthetic film is, in effect, one image continually transforming into other images: metamorphosis. It is the one unifying force in all of synaesthetic cinema. The notion of universal unity and cosmic simultaneity is a logical result of the psychological effects of the global communications network. (1970: 86)

The superpositional frames-within-frames aesthetic of The Pillow Book does not produce this conflict-free harmonisation of opposites. Rather, like the grotesque images of Greenaway’s dismembered bodies, they are disconcerting representations of a violent process. Indigestibility—and, with it, illegibility—is not the flaw but the achievement of the film. Eisenstein’s ‘precarious equilibrium (or, better, the dynamic tension) between unity and disunity, the centrifugal and centripetal’ is pushed to an extreme, as narrative bodies and formal elements declare their difference and display their interstices in intermedial synchronicity. (Tsivian, 2002: 49–50) The cuts are made starkly visible in a frame in which difference is performed, and not effaced by the ritualised repetitions of continuity editing conventions.

Even in an age of digital manipulation which permits pausing and replaying, the indigestible frames of The Pillow Book resist definitive reading. Daly argues that ‘[o]nce movies exist in a mode of replay and are part of online discursive space, the narrative can be more complicated and intertextual.’ (2010: 85) The superpositional structure of the frames-within/above-frames of The Pillow Book, though, does not permit enough interactive control to result in a more interpretable film. The frame moves, to invoke the celebrated essay by Susan Sontag (1966), beyond the possibility of interpretation, into what Julia Kristeva (1980) describes as the semiotic, where the foregrounding of rhythm, tone, mood, colour, and texture dominate and result in semantic imprecision. Hans Ulrich Gumbrecht outlines the ways in which ‘“interpretation” (and, with it, “expression”) [became] the predominant and, soon afterward, the exclusive—paradigm that Western culture made available for those who wanted to think the relationship of humans to their world’. (2004: 28) This dominance of hermeneutics, Gumbrecht suggests, is predicated on the parallel dominance of the Cartesian subject, whose ‘human agency in the production of knowledge became a condition for knowledge to be acceptable’. (2004: 34) The Pillow Book’s experiments in intermedial form do more than simply make the conventionally-invisible visible; they critique how reading-for-meaning has come to dominate theories of knowledge.

Dismemberment within The Pillow Book is the literalisation of a key metaphor (skin as book) and a key synecdoche (skin as body). Like the alinear dismembered body that no longer contains the centring properties of recognisable identity, Greenaway’s frame is a confusion of flat images synchronically arranged. This form resists orthodox reading practices. Gumbrecht argues that, from the Renaissance, there has been a dominant model of ‘world-interpretation through which the subject penetrates the surface of the world in order to extract knowledge and truth as its underlying meanings.’ (2004: 27–8) This celebration of modes of penetration (and of going beyond-ness, digging into-ness, transcending-ness) is ‘backed up by the positive value that our languages quite automatically attach to the dimension of “depth”.’ (Hans Ulrich Gumbrecht, 2004: 21) The Pillow Book stages an assault on this epistemological tradition by remaining—as its principal metaphor and synecdoche suggest it will—skin deep.

Skin-Deep Character

Much criticism, whether appreciative of the work or not, has sought to find a meaningful core to The Pillow Book, to stabilise the film’s superficial disintegration through a reconstruction of its unconventionally hewed and hacked body, to make sense of its (dis)arranged limbs. Serge Gruzinski and Deke Dusinberre insist that profundity arises from a predictable pattern in the film’s form, defined by its central character:

Greenaway employed various techniques to achieve this end. He selected and combined three distinct screens: a CinemaScope format, in color, when relating Nagiko’s life in Tokyo (the camera is placed at ground level, as in Ozu’s films); a smaller screen that recounts—in black and white—Nagiko’s earlier life and subsequent end; and finally, even smaller incrusted images of highly refined color, which illustrate passages from Sei Shonagon’s original Pillow Book like so many animated miniatures from that distant period. Instead of melding into the primary image, these precious scenes exist alongside it; the interconnections produce a series of visual effects that transfigure the action. This juxtaposition of periods sheds light on and validates Nagiko’s plans, while the refined colors of the past underscore or counteract the grayness of modernity. (2002: 81–2)

Even a cursory survey of the film reveals the inadequacy of this analysis. The representation of none of the periods discussed has the consistency of style suggested, and exceptions to these rules are as common as adherence to them. Instead, the film adapts its own formats endlessly, failing to provide the stable insights which Gruzinski and Dusinberre identify. The idiosyncratic film syntax is, as it were, descriptive rather than prescriptive, and the demands of an individual frame outweigh the importance of any overarching consistent system.

The film’s complex interplay of dismemberment and re-memberment cannot be completely explained away with reference to character memory. Nagiko recalls her father (Ken Ogata) and his exploitation by the publisher, and the death of Jerome, and these memories seem to drive her quest for revenge. However, the film is not quite willing to play ball. Character, like form, is performative, flattened into a series of enigmatic and monotonous performances, avoiding what Greenaway, in an interview with Marlene Rogers, describes as ‘that pseudo-supermarket Freudian analysis of character which becomes so boring’. (1991: 11–19) In the opinion of Elliott and Purdy, the critical effect of this eschewal of autonomous psychological expression ‘has tended to alienate those critics who define depth in terms of psychological identification.’ (2008: 247) Greenaway says ‘[t]he camera is stuck with perspective, but cinema no longer is; it can use frames within frames, or use special effects to shrink characters to proportions fitting to their status.’ (Paul Melia and Alan Woods, 1998: 29–30) The sliced and shrunken characters announce their own flatness in challenge to conventional narrative cinema which has privileged character. Nagiko becomes an encoded cipher producing encoded ciphers (through her on-skin writing) in a relentless mise-en-abyme; in her search for self-expression through text, she is, herself, textual, written on by her father and lovers and in a permanent state of homage to the earlier—and enigmatic—writer, Sei Shōnagon. The film’s characters are constituted by what they do and, in the mediated intermedial narrative world, by how they are shown to be doing it. In the restless and relentless pulling away and together of the dismembered and re-membered film form, Nagiko does not survive as a coherent self who can suture the intermedial mosaic. The film’s concentration on the disintegration of characters (both alive and dead) and film form suggests that only an act of reductive mythmaking can give the impression of stable coherence.

This approach to character sets The Pillow Book apart from Greenaway’s other works. Prospero’s Books also performs saturated frames of competing digital images, staging the unsettling of systems of meaning-making. The worlds of the two films are both performative, but the central characters occupy different roles; Prospero delivers every spoken line in Prospero’s Books, summoning into existence the complex culture of his world; his words do not remain inert on the pages of his books, but throb and gyrate and ultimately leap from their pages into the world beyond. Elliott and Purdy observe how ‘the metaphor of the playwright/filmmaker as magician/illusionist—with much deliberate cross-identifying between Prospero, Gielgud, Shakespeare and Greenaway—authorises a veritable orgy of framebreaking, affirming the fundamental continuity and interpenetration of book and body, image and world…’ (1992: 182) Prospero’s Books is centred by Prospero, Shakespeare, and Gielgud, whose charismatic presences hold the film’s shape. Nagiko—and behind her Sei Shōnagon—also breaks the frame, but she does not represent continuity between book and body; where Prospero’s words-on-the-page are animated as bodies, Nagiko’s words dismember and flatten bodies into pages until skin (figuratively or literally) is all that remains of them. The Pillow Book, with its emphasis on superfice and fragment, does not allow Nagiko to become a character.

Collage’s dismantling of chronology complicates conventional uses of psychologically explanatory backstories. When Nagiko as an adult waits to speak to the publisher, an inserted frame in the lower portion of the screen recalls the time when Nagiko as a child sat and waited for her father who was visiting the same publisher (Figure 6). In both frames—past and present—the office door opens at the same moment and the publisher is framed emerging with, respectively, Nagiko’s father and Jerome. The logic of this insert seems clear: Jerome is the successor to Nagiko’s father (both write on her and receive her love in return, and both are in sexual relations with the publisher), and the past, as the front layer within the frame, will always cover, eclipse, even obliterate the present. This seeming-logic implies that Nagiko’s remembering is what is re-membering (and holding together) the film’s discrete narrative limbs.

Figure 6

Uneasy past-present relations in The Pillow Book (1996).

However, this unidirectional temporal cause-and-effect relationship is unconvincing. The childhood memory should, according to the practice elsewhere, be black and white, but it is in full colour; the past here may cover the present, but the film, as it often does, complicates this causation and the present colours the past. The synchronic form of the frame insists upon a comparison of the two moments, but this insistence is futile. The comparison of Nagiko’s father and Jerome is an example of supermarket Freud: the former is seemingly coerced into sexual relations with the publisher while the latter welcomes them, and while the father’s destruction is attributed by Nagiko to the publisher, in the case of Jerome, her jealousy is a primary cause of his death. The simultaneity of the images invites interpretation—coincidence suggests correlation and even causation—but no interpretation can fully explain the relationship between the two frames. Meaning is contingent in the screen’s mise-en-abyme, caught between the distorted reflections of one frame in its neighbour. The parallels between the two events are, as Nagiko implicitly acknowledges, skin deep.

The film’s sense of skin-deep character is emphasised by the use of superposition. This effect can be seen clearly in the sequence depicting Nagiko’s final argument with her husband, in which the dramatic action is recorded in a yellow-tinted inserted frame, a top layer which gets bigger by stages as the argument proceeds. As it begins, the argument is contextualised by the hellish image of a whirling red fan flanked by two red cones. These items are remnants of the previous frame, a monochromatic car ride made by Nagiko and her husband. Suddenly vibrantly coloured, they are as much theatrical curtains and spotlight as roadside objects; they frame the argument, declaring its theatricality, both dramatic and artificial. This artificiality resonates with the performances of the film’s cast, the members of which throughout the film never convincingly inhabit its dramatic spaces, marking the defeat of the 3-dimensional fictionality of the narrative world. Kenneth Turan (1997), in sympathy with many of Greenaway’s critics, identifies this artificiality as a ‘dramatic weakness’, but this lack of depth marks The Pillow Book’s central formal conceit.

When the red frame subsides, replaced by a full screen frame of the samples of Nagiko’s writing which the argument is about, far from offering explanation for the argument, the image is gratuitous; it would be enough for Nagiko and her husband to say that the argument concerns her writing. The principal effect of the text, instead, is to flatten the images of the argument; the resolute 2-dimensionality of the blue written page resists any illusion of depth, a flattening effect enhanced by the centripetalism of the isolated front layer; the argument becomes an illustration on a page, a decorative design. The writing, illegible to many without translation, is a series of signifiers that demand and resist reading, promise and withhold meaning; this resonates with the semiotically perverse operation of all the film’s signs. The Pillow Book is concerned with skin and page, and its images (including text) are consistently prevented from going any deeper.

More complicatedly, the film does not present images of consistent, uniform flatness. Front and back layers are in an unstable dialectic, and mercurial changes in aspect ratio redefine their relationship. Within the argument sequence, a sudden shift to a CinemaScope ratio allows the front layer to make contact with the overall frame’s left and right edges. From there, with centrifugal energy, it can bleed out to the world beyond. Without the flattening context of an encasing back layer, the scene and its characters acquire the potential for depth. This potential seems to be within grasp as the back layer is lost altogether. Superposition is abandoned and, instead, the flames of the fire are superimposed over the continuing argument; the use of superimposition, rare in the film, creates a greater sense of depth through the suggestion of background and foreground. This, however, is only a fleeting and tantalising taste of possible meaning. Almost immediately, a new superpositioned front layer is introduced, an image of the flat (and flattening) book, now burning. The back layer (Nagiko’s diary) has simply swapped places with the front layer (the argument), even swapping aspect ratios. The burning book now seems flattened by the argument, whose staged quality continues to resist a reading of depth.

The utopian association of layering with depth—and depth with profound meaning—is exploded by the film’s flattening superpositions. The narrative’s suggestion of depth (as the past colours the present, as personal trauma informs behaviour, as memory resonates with experience) is compromised, and the possibility of meaningful character withers. The notion of synthesis between and within signs is repeatedly interrogated: the calligraph modelled in Nagiko’s writing is a partial collapse of image and text, and the overall frame is a mass of mutually-flattening frames. Text, character, and image—all divided—cannot be brought to life by film, and are destined to remain as flat as the pages of a book; knowable meaning and character, the film suggests, are always a myth. Taken out of the then-and-then structure of conventional psychological narrative that insists on explanation and even determination, the film instead suggests that marriage, love, sex, violence, divorce, and character itself are ritualised performances that are only meaningful because a culture—and often one that is structured on unequitable principles—declares them so. Characters can, however, assume a kind of vitality in the way that their constituent parts provoke each other and, while flat, they may, like the pages of a book, turn and transform. As Barthes suggests in The Pleasure of the Text, ‘[p]leasure [is] in pieces; language in pieces; culture in pieces. Such texts are perverse in that they are outside any imaginable finality.’ (1975: 51) The changing aspect ratios of the gradually engorged front layer in the argument sequence are both meaningless and suggestive, flat and mobile, indefinable and sensational. To hijack E.M. Forster’s celebrated appraisal of characterisation in the works of Dickens: ‘there may be more in flatness than the severer critics admit.’ (1990: 76)

Experiencing The Pillow Book

The question remains how to make the most of this flatness, how to find pleasure in pieces. Youngblood’s prophecy for the emerging synaesthetic cinema argues that in such a cinema, formed of syncretically aligned parts, ‘[w]e are not asked to interpret or find “meaning” in these combinations’ but rather to ‘experience our own involuntary and inarticulate associations.’ (1970: 87–8) These films offer, he argues, ‘an experience rather than a meaning.’ (1970: 126) This position beyond interpretation marks a key point in a line of experimentation in superposition, resonating with Quantum Theory’s understanding of ‘superposition’: the existence of linear structures in indeterminate states, the simultaneous existence of one structure in multiple conceivable positions, and a paradoxical indeterminacy only resolved into a static position at the moment of measuring. Youngblood argues that ‘because we’re enculturated, to perceive is to interpret,’ and to interpret is to assign values which must limit the indeterminacy of a text. (1970: 76) The linear states of The Pillow Book operate both synchronically and diachronically, both paradigmatically and syntagmatically, both integrally and disruptedly, and by simultaneously occupying multiple conceivable positions generate indeterminacy. This results in what Tweedie sees as an ‘explosion of the sign into the manifold aspects elided in the moment of inscription.’ (2000: 107) The explosion is both threatened by and a threat to the cultural codes which Youngblood fears enculturate us. As Paul Melia writes:

… direct access to material reality outside of representation is not possible: cultural codes allow us to observe not so much nature as nature which has been exposed to our method of questioning, classifying and ordering. Once again, Greenaway is parodying the assumption that the world can be completely represented in language… Here Greenaway denounces the Enlightenment’s faith—which, for many, has survived the twentieth century’s catalogue of disasters—in the possibility of both progress and a total, finite and rational domain of knowledge. In its place he offers us an epistemological nihilism: any act of understanding or representation is necessarily provisional, contingent, temporary, relative, incomplete, and will be marked by accident, chance and randomness. (Paul Melia and Alan Woods, 1998: 16)

While Melia is right that categories are dismantled in The Pillow Book, exposed as only the limiting evidence of ‘questioning, classifying and ordering,’ he is, nevertheless, too emphatic in his alignment of the films with ‘accident, chance and randomness.’ The beautifully composed frames insist that The Pillow Book is not the product of randomness. Rather, the film is the continual interplay of order and disorder, an interplay that is precisely a quality of dismemberment, a process through which a body both ceases to be and remains itself. Indeterminacy is not the same as ‘epistemological nihilism’; The Pillow Book is a proliferation of unmeasurable and concurrent states, linear knowledge reimagined non-consecutively, simultaneity reimagined as fragmentation, character reimagined as a ritual of performing selves.

At the core of this indeterminacy is the deployment of fetishized technologies, which become objects of desire and take the place of significance. Thomas Elsaesser sees this as leading to ‘post-films’ in which the ‘concerns are neither narrative nor iconic-photographic,’ a cinema which is ‘[i]n keeping with the new technologies it deploys, … graphic, concerned with trace and body, with surface, rather than space…’ (1996: 78) Spielmann is emphatic in her conviction that ‘the incorporation of digital effects into the technical means of photography and film enriches and expands the languages of these media.’ (2006: 60) This is not a dilution, still less a dissolution, but an enrichment, allowing new pinnacles of indeterminacy as the frame becomes an increasingly flexible space. Greenaway himself has frequently discussed the visual and intellectual possibilities opened up by this emerging technology:

So the film itself is very much a palimpsest of what’s happening now at the end of the 20th century with the fragmentation of the relationship between cinema and all the post-televisual medium: the CD-Rom, the Internet … French intellectuals have criticized the film, saying The Pillow Book is not a film, it is a CD-Rom. I could think of no higher compliment… [W]ith the video clip, with the use of the talking head, that continual change of perspective of time, event, idea, action and intended use of tense, there is a brand new vernacular language which is being developed day by day almost incidentally and accidentally, much as I suspect in the way that the early Japanese language was created by Sei Shōnagon. She was often accused, certainly by her contemporaries, for her excessive use of Chinese quotation. Television certainly recreates or reprises or “quotes” the celebrated so-called fossilized forms of cinema. Television, shall we say, takes cinema as the Japanese vernacular did the Chinese language of the 10th century. We have new languages that are attempting both to erode the old languages, but also to deliver like a phoenix, knowing that the new languages have to be a combination of the old and the new. (Lawrence Chua and Peter Greenaway, 1997: 28)

An indeterminacy—resulting from the complex negative dialectics operating within frames that are characterised by superposition—stimulates sensation until the moment it is measured and evaluated, at which point its flatness permits only supermarket Freudian interpretations of characters and their unlikely activities. The Pillow Book’s indeterminacies are not settleable through the related myths of character coherence, cause-and-effect narrative, syntagmatic form, or final meaning. The ideologies that underpin these myths are dismembered in the film in an autopsy of the medium that also offers a means for reviving the re-membered corpse. The film’s principal revenge narrative is the revenge on these myths that remain steadfastly foundational (the film does have recognisable character, a discernible plot, and identifiable themes) but inadequate as explanation.

This inadequacy can be seen at work after Jerome’s funeral, in the beautiful layering of superpositioned frames, which create an extraordinary almost-sense (Figure 7). The voiceover explains that the front layer is a burning effigy—a fake with direct signifying links to the original—of Jerome’s car. The independently moving middle and back layers remain unexplained; they are, perhaps, pages from Nagiko’s diary and the publisher’s bookshelves. Before they can be explained or meaningfully combined, they fade. This lack of determined meaning does not diminish the effect of the sequence. The divided frame is discontinuous and discordant, and yet the layers form an aesthetic harmony with each other. The vertiginous simultaneity of the saturated frame—a frame in sympathy and at odds with itself—teeters on ‘the verge of “countercommunication”.’ (James Tweedie, 2000: 108) Amy Lawrence sees this as ‘Greenaway’s taste for “sending out many messages simultaneously” to ensure “a diversity of multiple meanings”.’ (1997: 92) It should be seen as more than a question of taste, though. Bazin argues that ‘[i]t is in pulling the work apart, in breaking up its component parts, in making an assault on its very essence that [a] film compels it to deliver up some of its hidden powers.’ (1967: 169)

Figure 7

The almost sense of the dismembered frame in The Pillow Book (1996).

The Pillow Book is a constant yoking of disparate terms in a vertical montage which, without the binding of narrative linearity, exposes the cuts which both sever and re-member the image. As the narrative ends, Nagiko writes on her baby’s face, while an inserted front layer of her father (now in colour, both striking and jarring) hovers suspended, representing everything between Guardian Angel and vengeful fixation, between cathartic moving on and the relentless return of the traumatic repressed (Figure 8). The father does not mean deeply; he is, in effect, merely a mark on the baby’s skin. Nor does the frame mean deeply. Character and the frame are performed as combinations of parts—skin deep, dismembered, and indigestible—which, in Gumbrecht’s words, create an ‘”aesthetic experience” [which] always provides us with certain feelings of intensity’ and which move beyond ‘interpretation and understanding.’ (2004: 99) Intermedial films like The Pillow Book offer startling possibilities in the construction of an erotic cinema whose profundity lies in its demythologisation of the deep meanings of conventional film form, character, and hermeneutic criticism. In sympathy with The Pillow Book’s play with the medium’s performative reproduction of the illusion of meaningful structures, critical responses should, perhaps, focus less on what the film means and more on what it does and undoes.

Figure 8

Skin deep meaning in The Pillow Book (1996).

Competing Interests

The author has no competing interests to declare.