Brakhage’s Blacks

This is a version of an essay published in Stan Brakhage the Realm Buster (John Libbey press, 2018, isbn: 978-0861967285. The printed version carries an error on page 4 that I have corrected here.

 

“The speed of fade and the time length of the black reminds us that movies aren’t moving pictures only: structurally, they’re time-based graphics (like a black screen), some of which aren’t pictures at all”1.

“Black has the specific quality of being only ever virtual. Natural luster, imperfect pigments, ambient light, and neighboring colours all inflect surfaces we perceive as black: achieving solid, lasting blacks takes considerable effort, the more so the more we deal with screen media that either reflect or emit light as the basis of their working” 2.

“There is metaphysical pressure to keep the contribution of shadows ‘off the books’. Philosophers and physicists alike have a strong conviction that reality is positive. They think a negative statement such as “There is a permanent absence of light in the Shackleton crater” is really about where light is rather than where it is not…(they) are uncomfortable with absences and so gerrymander discussion to disenfranchise black shadows, black space and the black sky of the lunar day”3.

 

The ideas for this essay have their origins in an earlier piece on Brakhage’s Roman Numeral Series4, which touched briefly on the various ways Brakhage used black in those films. Subsequently the work was developed, rather haphazardly, in the presentation at Birkbeck, of which this essay is a refinement. In a lot of thinking on the topic, black is conceived as negative, as an impure absence, but my aim here is to show how the blacks in Brakhage’s films, while they are literally areas of apparent absence of light, or at least relatively reduced absences i.e., they can be construed as ‘negative’ in this strict sense, they are also, or also function, positively, in imagistic, graphic and structural terms. For painters, black is conceived of as a colour, something that is evident in its availability in a variety of shades with exotic names: Lamp, Ivory, Mars etc. These pigments reflect varying quantities of light, and each has a colour-cast, i.e., is impure. A truly black black, one that would absorb almost all of the light cast upon it, exists. Vantablack absorbs 99.96% of light.5 Unlike the paints and the film blacks described by Sean Cubitt above, which are subject to the influence of ambient light among other things, Vantablack absorbs everything, and this puts it strangely out of balance with other elements when it is combined to make a painting or other kind of image, because it falls so far outside the typical contrast ratios of a painting or a photographic image of the World.

Black, and the shadows with which it is often associated, have a curious status in film. Notably in Films Noir, shadows are often completely solid, which is almost never the case in perceptions of the real world, where the eye can peer into relatively solid blacks and adjust to differentiate detail and variety of density. Shadows in Film Noir have a quasi-autonomous compositional function and thus become a structural part of the image, imparting a degree of abstraction to what is otherwise a representation, but this is rare in cinema generally.

Cat People

Cat People, Jacques Tourneur, 1946.

Whereas black in a painting is a reflective coloured surface, and is what it is in its literal material sense, in the celluloid filmstrip it has a curious double status. At the point of image formation, ‘black’ is simply the area of the film where light registers strongly on the film’s light-sensitive surface. Materially it is an area of density (silver halide crystals that have been blackened by light) that holds back that light in projection, so is in a sense a refusal or interruption thereof, while at the level of the image it is, paradoxically, a representation of the (relative) absence of light, i.e., shadow or darkness. More precisely, it defines an area where there would be visible things had there been enough light when the image was made. Yet insofar as this absence impinges upon other areas of the frame contents, it becomes part of the image’s form: it takes up space in the frame. Furthermore its density means that it is often stronger –denser- than other parts of the image that in its pro-filmic would be materially more substantial: the image of a shadow cast by a tree may be stronger than that of the tree itself.

The question of whether black is an image is also dependent on the specificities of a film and how it is conceived. Films like Peter Kubelka’s Arnulf Rainer (1960) and especially Tony Conrad’s The Flicker (1966), which are composed of wholly black or white frames, are conceived as imageless, because the intention is to create patterns of light interruptions, structured from the presence and absence of light on the screen. In the case of The Flicker this was achieved by exposing the film to light with a lens-less camera, and by taking frames with the lens cover on for the white and black frames respectively. The filmmaker Bruce McClure creates his black and white loops simply by bleaching selected frames on a strip of black leader to create the white frames, while leaving the black frames untouched. In these examples there is no intention to create an image and thus no image in the usual sense, even though the flicker stimulates hallucinatory colour patterns when the films are viewed. A selection of Brakhage’s films can be made that demonstrates all of the above, including the last category of imageless blackness, which is a feature of Passage Through: A Ritual (1990, colour, sound, 50 minutes) and which is discussed below.

The creative use of black has been a feature of Brakhage’s oeuvre, starting right from the beginning with the noirish psycho-drama The Way to Shadow Garden (1954), a high contrast black and white film that turns to negative in the second part, after its protagonist has blinded himself and become a seer. At the end of the opening shot the camera pans to settle for five seconds on the exterior of the house in which the film will unfold. Dazzling light pours from two square windows, which are framed by an entirely solid area of black, creating a simple abstract composition, something which establishes a pattern for the rest of the film, as well as, in a wider sense, Brakhage’s working against the simplistic dichotomy between representation and abstraction.

WTSG

WTSG.Bulb

A single bare bulb is established as the film’s only apparent light source, giving us to suppose that this precarious illumination is all that’s keeping the film alive, warding off the total darkness that would otherwise ensue. At one point the camera lingers on the bulb in close up, seemingly in an effort to break with the positive meanings usually associated with light. The oppressive light momentarily threatens to obliterate the image.

The young protagonist lurches around inside the house and we see, inter alia, a framing with a black wine bottle on a table on the left side of the frame, his shadowed face on the other, thus a graphic presentation of two strongly contrasting forms of black. At the point where, having blinded himself, he staggers towards a French window and opens the doors, the composition is divided into three roughly equal vertical bands. The man is framed centrally in silhouette, bordered on either side by black. At this point the shot cuts to negative and in what were solid black borders we abruptly see previously invisible details. Where the positive part of the shot is back-lit to create a silhouette of the man, the following negative version is front-lit, revealing details in the image where there previously were none.

WTSG.4

WTSG.Neg

This framing is pre-figured earlier in the film when the camera settles momentarily on the same pair of doors, between which is a rectangle of solid black, thus a reversal of the disposition of black and white in the two shots, a move which also prefigures the later transformation into negative. A similar reversal also occurs when the windows seen from the outside in the opening shot are seen from the inside, before the protagonist desperately lowers the blinds on them.

WTSG.2

WTSG.3

After he has blinded himself he picks up the lamp and waves it around, causing shadows to play on the walls. He then puts down the lamp and writhes around between the lamp and a white wall, continuing the shadow play by other means.

WTSG.3a

WTSG.Shadowplay

 

Anticipation of the Night (1958) is the film that marks Brakhage’s transition from human-centred psychodrama to a corresponding form in which the camera replaces the figure as protagonist. It opens with the same motif as the transitional shot in Shadow Garden, but framed very differently: an illuminated threshold, bordered by black, across which the shadow of a figure passes, momentarily darkening the screen. As if to mark the transition from rectilinear framing to the more oblique angles that will figure in subsequent films, the borders of the threshold fall diagonally across the screen, fanning out in a manner that strongly evokes the conical shape of a projector’s light beam or indeed the cone of human vision.

Anticipation

This was the point at which Brakhage rebelled against the conservatism of conventional optics and stable points of view modeled on traditional perspective, in favour of a subjectivised vision achieved by hand holding the camera. In this sense the sequence may indicate a kind of farewell to conventional framing. The shot is repeated several times in the  opening minutes, first from one angle then the opposite – not the reverse, as in narrative shot-reverse-shot grammar- but laterally inverted –mirrored, probably by flipping the negative over during the printing stages. The shot becomes a mobile movement of black within a light frame, from representation to abstraction, breaking down the distinction –which Brakhage disliked- between the two. This figure in a doorway is interspersed with compositions that contrast strongly in a formal sense, but which are made of similar stuff: sparkling points of white light on a dark ground. In both though, there is a flattening of the image through the silhouetting process, and this continues in twilight shots, filmed contra-jour from a moving car, of trees, which obliterate the slivers of evening sky. There is also a repeating pan round an object, which in effect functions as a horizontal wipe, from light to black. These shots also prefigure or anticipate (sic) the onset of night and thus black performs a precise semantic-thematic function.

As the film moves from day to night there is a form of double reversal from negative to positive and vice versa. In the first sense the trees’ obliterating the night sky is a negation through interruption (a rhythmic process that also figures the way the projector’s shutter interrupts the light at the moment when the next frame of film is pulled into the gate to be flashed onto the screen) of the illuminated scenery. These early scenes can be seen as light fields interrupted by dark movements. In the second sense there is a reversal of this field, from light to dark, against which spots of artificial light, and sometimes the moon, assert themselves uninterruptedly against the blackness of night. In their assertiveness they present as positive in a way that was not the case in the twilight part of the film.

Insofar as Wedlock House an Intercourse (1959) features figures –Brakhage and his new wife, Jane Collum- it returns in part to the early psychodramas. At the same time it is one of his blackest films, with long sections where the screen is entirely dark apart from a small point of light or a curved shape arcing through the frame, though the arcing effect is as much that of the lamp’s movement as it is the movement of any object. Brakhage makes much use of this device. As in Shadow Garden, the illumination is precarious and highly unstable, a hand-held lamp with a naked bulb and occasionally candles –raw lights- being the only apparent sources. The lamp is rhythmically waved to create flashes of bleached highlights and enveloping shadows on and of doorways, often reflected in mirrors, creating a double framing. Cigarette smoke, both frontally lit (white) and silhouetted (black) and an alarm clock, interspersed with repeated sexually explicit shots in negative, are rhythmically permutated. The film is notable for the way in which what is thematically an explicit film about young love is rendered in so highly an abstract way. The blacks are dense and solid and the fragmentary scenes emerge sporadically out of this dark. Black consistently dominates the image, immediately enveloping the momentary light flashes, so that rather than functioning as part of a differentiated pattern made from a range of grey tones, as it would in a negative of a conventional image, black and white are in battle.

 

As Brakhage’s career develops one can detect a gradual move towards a concentration of colour and crystalline forms, which are often expressed as highly saturated colours, both primary and prismatic, as if the more broadly distributed colour of many of the films made in the 1960s and 70s condense into small intense areas. In these works, notably films like Arabic 12 (17’, colour, silent, 1981), black becomes more like the colour described by Cubitt above, inflected and tinted by adjacent light spilling into the dark areas through refraction, reflection and lens flare.

Arabic 12

Arabic 12

The blacks in this film are less dense and often very grainy, sometimes because the prints are blown up from Super 8 originals. Often the black grain briefly freezes to form a greyish textured surface, upon which, or seemingly within which, new colour movements develop. Although, then, the film has been reworked in an optical printer, it has the quality of an improvisation. The hand-held camera is pointed at the sun through indiscernible obstructions, through and around which light leaks into the camera in the form of prismatic flashes, lens flare, including isolated arcs, lines of intense, star-shaped colours, and curvy, deformed geometric shapes; rhomboids, rectangles and triangles, as well as less defined patches and hues, including a bluish cast that overlays the whole image at times, modifying the contrast unpredictably.

The filmmaker juggles with these elements, shifting the balance by increasing or decreasing the amount of light / colour in relation to black and by pointing the camera directly and indirectly at the light source. The object causing the blackness isn’t itself black, but a blue something that most of the time is held too close to the lens for light to reach it. However it more often than not appears black. It has a kind of double status: any object will appear black when it’s under-lit, as is the case here, but it also generates blackness by blocking the light to the camera, thereby depriving itself of light. The moving colour shapes also have movement occurring within them in the form of animated texture, and this texture impinges on the notional black surface too, so that its own colour changes. There’s a gesture to Goethe’s colour phenomenology, in which the prismatic colours arise in the interaction between white and black, which is appropriate because black here does appear to contain and juggle many colours. It is chromatically mutable, assuming surrounding colour-shapes, squeezing and dispersing, intensifying and darkening them, catalyzing an endless succession of transitory phenomena. Throughout the film there are freezes, at which point the black becomes a static, textured grey field. Almost immediately small movements begin within this field, but the freezes remind us that underpinning the image is a volatile ground of grain movement, which becomes most visible when it is seen in greyish mid-tones (grain is imperceptible in areas of pure white and black). In Paul Sharits’ Axiomatic Granularity (1973), grain is re-filmed and magnified to the point where the solidity of a plain colour field fragments into dancing crystals of grain, illuminated by the light that strikes through and around them. Thus the solidity of any given colour on a film strip, including, black, is dependent in part on the density of grain and the magnification at which it is observed.

Arabic 12 is one of Brakhage’s most restless and probing films. It is the product of an interactive interplay between hand, eye, camera body, lens, film, light and objects. The lens’s limitations and weaknesses are harnessed to generate an extensive range of optical events that also echo Goethe’s interest not only in: ‘subjectively produced colors (after images, light and dark adaptation, irradiation, colored shadows, and pressure phosphenes), but also in physical phenomena detectable qualitatively by observation of color (absorption, scattering, refraction, diffraction, polarization, and interference)’6. Refraction was crucial to Goethe’s account of colour phenomena, and Arabic 12 is replete with both refractions and diffraction in the form of prismatic fan-shapes that arise seemingly out of nothing.

At the same time as making the Arabic Numerals, Brakhage also worked on a smaller but similar series of seven films, the Roman Numerals (1978-81). Like the Arabics, they eschew the distinction between representation and abstraction. The films are ineffable  -hard to write about, appropriately, given Brakhage’s concept of film as ‘moving visual thinking’, and hence antithetical or at least resistant to exegesis. As in Arabic 12, Brakhage often appears to be working with a set of elements, if not a formal system, in which parameters like focus, brightness and depth of field, and occlusions, like the unidentifiable black object of Arabic 12, are interplayed through camera movement and aperture manipulation during shooting, though in a manner more restrained and less rhetorical than in Arabic 12. Most of the films in the series are blown-up from Super8, hence the heavy grain structure generates pointillist quasi-separated colours, which are an integral part of how the films work: they are colour films in the same way that there are colour paintings (Derain, Matisse, Newmann, etc)

 

Roman I.

The first film (1979, 6’) has a predominant scheme of off-white through salmon pink to deepish mid-red. A camera circles over a knot of fuzzy, reddish lines. Underneath (or beyond?) these light, caressing moves, presumably lies an unidentifiable object, but it is just as easy to see this knot as a spatial array, since it is neither enclosed by a contrasting border which would locate it in a putative space, nor does it appear to be cropped by the camera’s framing. As such it has no perceptible depth but neither does it obviously lie on the picture plane. The familiar octagonal refracted image of the lens diaphragm, formed when light enters the camera directly, are rendered circular by the lack of focus, and form an integral part of the image. Could this be because the object is a light-cluster -incident light- and not an object, which would help it to harmonise with the light/lens refractions, or is it simply that defocusing blurs, literally, the distinctions between objects and light, or rather between reflected and incident light? Here are the first of the many cinematic antinomies that are dissolved in the film, between the pro-filmic and the apparatus, between conventionally wanted and unwanted optical phenomena.

RN1.2

Periodically, we see what appears to be a zoom, but into what? The idea of zooming supposes a final target detail in a predefined field, but when that field is already undefined -defocused- there is little against which to measure the zoom’s progress, and thus the distinction between wide and close loses its purchase. What we have in effect is a kind of reframing, or better still, a pure cinematic movement, one that is not dependent on a pre-established pro-filmic which retains itself as a function of the apparatus, but a new kind of abstract movement which progresses, or evolves, the image.

 

Like Arabic 12, many of the Roman Numeral films consist of prolonged shots, in which Brakhage gives himself time to establish and develop a similar juggling of elements, though the films are generally more enigmatic, less concretely and discernibly of something, and, by contrast, more luminous. Single black frames perform a variety of functions in the film. They punctuate these longer shots, and dynamise the relationship between longer and shorter ones. Because the longer shots are often very similar to each other, the black frame serves to indicate the end of one shot and the start of the next by a clear and emphatic, albeit momentary, pause. The black frames also generate rhythm, or in some cases augment an existing one. They have a mildly disruptive effect, without creating pauses, as longer sections of black would. This disruptive effect can also be seen as an anti-montage device: once shots are separated, even if only by a blink of black, montage is frustrated, the clash of the cut which generates Vertov’s essential Interval, is disabled. It’s also worth mentioning that a single frame of white would be much more disruptive than black because it’s an intrusive, dazzling flash, as opposed to a moment of darkness. Therefore, the black functions in part as a discrete but powerful punctuation mark, which inflects meaning and rhythm in the same way a comma does in a text. These punctuations energise flowing passages of what are often closely similar shots.

RN1.4.strip

Roman Numeral 1

 

There is another way that black frames work, which contrasts with the examples given in the foregoing paragraph. It is quite often the case that the black frame is next to the over-exposed first frame of a shot (this can be seen in the frame strip from Roman 1, above). This happens because the first frame to be exposed receives more light than those that follow, because the camera is not yet running at full speed –it takes one or two frames to accelerate to 24fps, hence the first and to a much lesser extent the second are exposed for longer than subsequent frames, when the exposure time is stabilized at 1/50th of a second. This technical fact results in a visible flash when the film is viewed at normal speed, but the insertion of a black frame increases the contrast between the light and adjacent frames. By the same token this contrast also increases when the frames adjacent to the black frame are lighter.

 

There is one more thing to say about the insertion of single black frames between shots. If two pieces of negative are joined together using a cement join, the splice is highly visible when it’s printed, as the edges of the film overlap each other within the frame area. A simple and economical way to avoid this is to insert a frame of clear film between the two shots, which becomes black when the negative is printed. It’s also cheaper to do this than to use the special methods required to create invisible cuts, because laboratories charge less for printing from a single roll of negative. It’s quite possible that this was Brakhage’s reason for using this technique, but in any case it’s a good example of an economic necessity generating aesthetic effects, in the same way that financial constraints often prompted production of the many films he made by working directly onto the film strip without a camera.

RNV.1

Roman V

V (1980, 2’ 36”) has a roughly symmetrical structure, beginning and ending with greenish dark forms, with a long stretch of black in the middle, which is spotted with warm-toned flashes of colour. We also see for the first time shapes that are apparently generated from pro-filmic objects, in other words things which don’t look only like light effects, although, as always, one cannot be entirely sure. These too are dominated by black, so that the light parts of the image are strictly isolated. The main image is a kind of eclipse-like semicircular line, which partially encloses a roundish shape in centre-frame, before dissolving away:

RNV.2

Roman V

Black is articulated by colour rather than the opposite, as might be imagined, with many shots that are almost entirely black. Whereas in Arabic 12 a black object is used to generate and control colour phenomena, here the colours frame and hence define the areas of black as something, by implication.

RNV.7

The colours in Roman VI (1980) are green, orange and pale red: not pink, but a de-saturated red. Immediately one is set to think about the relationship between hue, brightness and saturation, factors that are in turn complicated by the brightness of the projector bulb and variations in the colour temperature of white. Pinkness is a function of white being added to the mix, and which, therefore, given that white is opaque, is an opaque colour itself. But the whiteness in a film pink comes from the projector lamp shining through a certain density of redness, so how is a colour like pink even possible in film? Furthermore, opaque colours tend to look flat, that is, non-shiny and so relatively homogeneous (think of Vantablack, which is absolutely homogeneous because wholly non-reflective). Yet in films there is always an impurity of colour, and a certain depth at the grain level, since the image is a composite of coloured layers in grain, as seen in Axiomatic Granularity.

 

The film opens with a brilliant pulsating soft form, a warm-toned and glowing shape that hovers over a field of dense black.

RNVI.3

There are long black pauses whose colour flare-ends fall within the frame. Thus each black stretch is announced with a momentary flash of brilliant red/yellow fringes and ends with the same, creating a kind of overhang into the next shot.

The sections of black spacing announce themselves as shots, not just pacing devices, through these colour flashes. They imply that the black has some kind of meaning beyond its rhythmic/pacing function. This may be nothing more than the intention to dramatise the contrast between black (colourless and lightless) and red-yellow (coloured and light). In this respect it relates to the single black frames in the first film. The sense of high contrast is continued within images, as opposed to between them, and the colour end fringes of the black sections are echoed in the colour fringing of actual images.

The colour fringes also generate their own optical effect, in a film of optical effects, and remind us that buried in black celluloid are colour layers which may only become visible if the emulsion is scratched into, or if the shot is fogged at its very end. Thus it is implied that black contains other colours, even if they may be be visible. They are latent, waiting be revealed, and once we understand this we can no longer think of the black as simply the absence of light or colour.

flare end

Roman VI

 

It is almost didactic in the way it works through a series of colour and graphic possibilities, starting with cuts between loosely complementary coloured, contrasting forms.

RNVI.3a

Roman VI

These are followed by a series of shots in which the interaction between different kinds of graphic forms and the constant tonality in the film, black, are rehearsed; slashes, a circle in lower right corner, diagonal divisions of the frame.

RNVI.4

RNVI.5

RNVI.6

These different forms all imply different kinds of black and all function in a slightly different way, even though the black itself doesn’t change. It doesn’t move, but it punctuates, articulates and dramatises, as well as unifying in a thematic sense, in that all the Roman Films are unified by a dominant colour. Black forms a background in that the lighter shapes imply it as such, but insofar as there is no depth to the black it is more like the ground in a painting than a background in the photographic sense. One wants to read it illusionistically as deep space, but the urge to do this is based on assumptions, which are themselves based on conventions, not on strict observation and reflection thereon. VI can be seen as a condensed, abstract echo  of Anticipation of the Night in the blance between light and black forms shifts towards the latter as the film progresses.

VII (1980, 4’ 45”) begins with what appears at first to be a TV roll bar, but it soon becomes apparent it is not: its movements are too discontinuous. It’s actually a black bar on a white ground whose position shifts arbitrarily, yet there is surely a comment on TV here.

RNVI.7 copy

Roman VII

Brakhage’s dislike of both broadcast TV and Video as media have nevertheless inspired some of his best films, including 23rd Psalm Branch (1966-7) and Delicacies of Molten Horror Synapse (1991), with its extraordinary ghostly grey textures and images in negative, overlaid with looped, hand painted material. In the Roman film the comment seems to be more on the violence of the electronic process by which the interlaced TV image is constructed and disseminated, a disturbing, a-rhythmic electronic emission.

Diminishing the size of a camera lens’s aperture extends the depth of field -the spatial axis between the lens and infinity- so that “stopping down” brings things into focus that were out when the aperture was wide open. Stopping down concomitantly darkens the image as progressively less light reaches the film.  Brakhage makes repeated use of these principles, creating fades to near black from out of focus beginnings, so that shots come tantalisingly into focus just as they fade away. One says “fades to black”, but in fact the device as used here is not really a way of fading, more a stategy for shifting emphasis within the image, in effect altering the hierarchy of elements within the scene, or “leading the eye around”, to borrow a phrase from painting. Darker areas, of course, are the first to go, and highlights the last, so that at a certain point in a given shot, the image will shift from being a variegated, multi-coloured field, to a dark, modulated one with a constellation of soft highlights. Here it seems black is the product of a technical procedure, conjured into being as the lens aperture is closed. An image is formed at the same time as it is destroyed, at the point that most of the frame goes dark. The last fifty seconds comprise a gradual darkening of a blue ground, so that one is again led to understand that what becomes black isn’t really. It is also yet another example of the application of Goethe’s idea about blue being next to black.

RNV.9jpeg

Roman IX

In IX, the final film in the series (1980, 1’ 40”), black spacing comes to dominate even more. The film begins with 17 seconds of black, which are interrupted by two half-second bursts of white. There is a similar vignetting process to Arabic 12 with star-like forms emerging and receding and the image of the setting sun emerging repeatedly out of a black field. The film has a “hot spot” at its centre, as if it has been re-filmed using a crude back projection system. Focus-pulls are deployed extensively, notably to change the shape of clusters of star-like highlights, which morph into horizontal bar shapes and back again, hence motion and mutation are achieved through focus pulling. An alternative approach to such cinematically created movement, that is, movement as the product of camera strategies, not pro-filmic movement, is rigourously elaborated in Wilhelm and Birgit Hein’s film Structural Studies (1970). Where the Brakhage films are restless and animated, the Hein’s is cool and methodical. It may be compared, in its taxonomical rigour, to the photographic work of Bernd and Hilla Becher, who photograph industrial buildings in a meticulously uniform manner. This is not to say that their work is not also about the apparatus in important ways, but there is a settled equilibrium in their pictures between the apparatus and its subject which in Brakhage’s films is always precarious and unstable.

 

In complete contrast to the restless and mutable black of Arabic 12 and the Roman Numerals is the bold and unvarying density of Passage Through: A Ritual (50’, colour, 1991), less than a minute of which has any imagery. It is one of Brakhage’s most uncompromising and experimental films. It falls far outside the terrain of the family romances, landscape studies and light plays that make up so much of his oeuvre. By contrast, Passage Through is surely the most austere and unusual film to be made by Brakhage, or indeed anyone else. It arose after he received a sound work by the composer Philip Corner: Through the Mysterious Barricades Lumen 1 (After F. Couperin), a long, minimal solo piano improvisation based on one of Francois Couperin’s most celebrated keyboard works, the rondo, Les Barricades Mysterieuse, the fifth piece in the sixth suite  –Ordre– from Book Two of his Pièces de Clavecin of 1717. The piece is pitched relatively low on the keyboard, mostly in the lower register, i.e., below middle C, and this is reflected in the score, in that both hands are written for the bass clef, instead of the standard arrangement of bass clef for left hand and treble for right. This gives the work a ‘dark’ tonality compared to pieces that are more equally balanced between bass and treble. When played on the harpsichord the piece has a rich clangorous quality, which Corner’s composition replicates in pianistic terms through long sustained accumulations of notes. Couperin’s short composition (two minutes to Corner’s 40 plus) is mesmerizing in its use of rotating figures in both left and right hands, among which repeated single notes ring out insistently, and Corner’s extended version is structured around repetitions derived from the Couperin, specifically the minor second that comes at the end of the first section of the rondo and which is repeated in the final. The minor second recalls some of Morton Feldman’s later piano works, notably Palais de Mari (1986), one of his very last compositions. Corner’s composition features rolling, repeated notes in the treble, which are periodically joined by noisy clusters in the bass. The sustain pedal is held down throughout (a device also used by Feldman in his longest work for solo piano Triadic Memories (1981) which lasts at least 90 minutes). After about 33 minutes of what is often nothing more than a single note played repeatedly, we hear the whole of Couperin’s original, played at about half speed, including the last two notes that form the minor second, from which Corner’s piece then directly resumes, making explicit its derivation from its source, and accelerating for the last few minutes of the film7.

Brakhage’s film challenges what a film is, or needs to be, or to have, to qualify as such and in this sense is almost para-cinematic8, in that it raises the typical question of whether a roll of imageless black leader running through a projector is a film: what do we see when we experience such a work? There is, on the other hand, just enough visual material to secure its being as a film, assured by the presence of the very few, tantalisingly short shots dispersed throughout. The first of these, a light source partially occluded by foliage and / or some kind of mesh, filmed through an out of focus window frame, appears after several minutes and lasts just over one second. Subsequent shots of similar length flash up every few minutes until after 25 minutes we see a relatively slow and lengthy shot of a cloudy sky that fades in and out in around five seconds, followed in short succession by shots of a similar length, before returning to longer periods of black.

Insofar as there is nothing on screen for most of the time, we have the space to listen to the continuous and rhythmic soundtrack, whose repeated notes are connected by sustained drone-like tones. The inevitable division of attention occasioned by the simultaneity of sound and image in most movies, something Brakhage objected to, and which is reflected in the fact that all but about six of his 450 odd films are silent, is avoided by the same logic in this case. The situation of, in effect, listening to a piece of music in place of watching a film is arguably the most conducive possible situation for so doing, more than at a concert, where there are plenty of visual distractions. Black is the ideal colour for the film: non-symbolic, non-associative, tonally neutral, devoid of any inherent modulating or irrelevant aspects (notwithstanding, of course, the variations due to different screening venues, levels of ambient light in the space etc). Equally, black maximizes the impact of the images when they do appear, by virtue of its contrast. The images are sufficiently infrequent so that one becomes immersed in the music, then, when an image does appear it is always by surprise and has disappeared before one has had time to take stock. Only at the halfway point, where the aforementioned cluster of slightly longer shots appear in quick succession, does one begin to get the measure of things, anticipating the next shots after two or three of about six have appeared. However, this is a film, not a static image –a projected slide, for example- and so one is inevitably also always in a state of anticipation, however much this is etiolated by the scarcity and infrequency of images: something, anything, could happen. On the other hand, the black is consistently solid and unvarying, so that the question of what we are looking at and why inevitably arises. The state of anticipation gives way to absorption in the music, or perhaps the mind, and indeed the eye, wanders. One might not look at the screen yet still be absorbed in the music, which is, after all, part of the film. A film without sound is unambiguously a film and no less so for being silent. On the other hand a film with sound but no images is a much more dubious object. What is it really? One is reminded of Clement Greenberg’s statement about the minimal requirements for something to be a painting, based on his idea that flatness and its delimitation were fundamental to painting: ‘thus a stretched or tacked-up canvas already exists as a picture, though not necessarily as successful one. (The paradoxical outcome of this has been not to contract, but actually to expand, the possibilities of the pictorial)”9. Thus, like the unpainted surface of the canvas, the black of A Passage Through is intended, bears intentionality and functionality, and this is true of almost all of Brakhage’s uses of black. Black is never simply or only a pause. It always has some other function, disruptive or rhythmic, or to serve a specific graphic purpose by contrasting with or emphasizing the character of adjacent shots or frames.

Brakhage has stated that Passage Through “required the most exacting editing process ever”. It is hard to work out what this might mean, however, at least in considering it at the level of sound-image relations in terms of synchronization. The main cluster of shots, a pivotal point in effect, occurs halfway through the film, but several minutes before the recital of Couperin’s rondo, which is a corresponding point in the music, however the relationship between the sound and the images –fleeting close ups of foliage, human gestures and skies- is impossible to pin down, and surely this was intentional, since anything that even hinted at some form of synchronicity would nudge the viewer to recuperate the work as a kind of quasi-generic music video or simply a conventional ‘film’, albeit of an extreme and unusual variety.

The brevity, semi-abstraction and tentative character of the images are wholly contrasted to the transparency and literalness of the music: like Feldman’s later works, the structure is on the surface, lucid, legible and almost didactic in its workings-out, unlike a lot of complex C20 serial compositions, where recondite analytical procedures are required to uncover the organizing principles of a work, work that furthermore made a virtue of avoiding repetition, thereby frustrating the memory functions that commonly help the listener to orientate themselves. For most of the time, Passage Through does not require memory function, since there is almost nothing to remember, on account of the limited and repetitive nature of the material. However, it makes acute demands at the point where images do appear and overall on reflection, at the end: what were those images and how are they related? Is there a cumulative effect, or are they deliberately spaced far apart so as to frustrate efforts to concatenate meaning?

The inflected blacks described by Sean Cubitt in the opening quotation are ubiquitous in Brakhage’s oeuvre, but there are a few films where they are explicitly explored. In The Worm of the Cat’s Green Realm (1997), a black cat is filmed washing itself, the camera zooming in until the frame is filled with iridescent, animated fur. These shots are contrasted with close ups of saturated greenery, behind and among which lies a density of blackness. Texture has migrated from the black of the cat’s fur to that of the leaves, concomitantly shifting the kinds and effects of black from one to another and so on. Almost uniquely, Brakhage’s films, in the process of getting us to attend to the complexities of our seeing the world, as inflected by the interplay between the myriad of perceptual phenomena the films both record and stimulate and the peculiarities of the apparatus, lead to a reflection on the possibilities of black. There is a negation of the negative, so that although black is always in some way the absence of light, this absence is never absolute, invariable or reduced, even in Passage Through, where its reduced function serves a specific purpose and is in any case crucially rescued by the few short shots it contains. Rather, Brakhage is constantly working black in different ways –formal, semantic, phenomenal, metaphorical- to diversify and expand its possibilities. Black is an equal partner in so many of these films that are more usually described in terms of their light qualities.

 

Thanks to Emile Vergé and Fred Camper for suggesting some of the films that I would not otherwise have other known about, and to LUX, London, for kindly making their collection available for me to view.

 

  1. Durgnat, Ray, A Long Hard Look at Pyscho, London: BFI, 2003, discussing the fade to black at the end of the first part of Psycho, when Marion Crane’s car sinks into the swamp.
  2. Cubitt, Sean, 2014, The Practice of Light: A Genealogy of Visual Technologies from Prints to Pixels, Boston: MIT (Leonardo Book Series), p. 22.
  3. Sorensen, Roy: Seeing Dark Things, New York: Oxford University Press, 2008, pp 16 & 201.
  4. Hamlyn, Nicky, 2005, “The Roman Numeral Series”, in Stan Brakhage: Filmmaker, ed., David James, Philadelphia: Temple University Press, pp 113-128.
  5. Surrey Nano Systems: Vantablack: https://www.surreynanosystems.com/vantablack
  6. Judd, Deane B. (1970). Goethe’s Theory of Colour, Cambridge: MIT Press, p xi.
  7. Les Barricades Mysterieuses has been a major resource for Corner in his improvisations. He has recorded improvisations based on it several times using a variety of similar titles, in solo and collaborative versions of differing lengths.
  8. Paracinema, a term coined by the American filmmaker Ken Jacobs, applies to films made without film technology –cameras etc- or to works that are called ‘films’ but which aren’t in any normative sense, for example Anthony McCall’s Long Film for Ambient Light, which was simply an empty room illuminated by daylight during the day and light bulbs at night. For an evolving theorization of Paracinema, see Jonathan Walley, “The Material of Film and the Idea of Cinema: Contrasting Practices in Sixties and Seventies Avant-garde Film,” in October 103 (Winter 2003), pp 15–30 and subsequent formulations. For a detailed discussion of Walley’s evolving formulations, see Cathy Rogers: Film Outside Cinema, (MPhil, Royal College of Art, 2014).
  9. Greenberg, Clement, 1993: ‘After Abstract Expressionism’, in The Collected Essays volume 4: Modernism with a Vengeance, 1957-69, ed. John O’Brian, Chicago: Chicago University Press, pp 131-132.

 

 

 

Outside Meaning.

This essay is an attempt to work through some of the problems arising from recent artworks in which the acquisition of specific information not contained in the work itself, nor to be confused with context more generally, is required on the part of the observer in order for the work to have its desired effect. It is a work in progress, so comments welcome.

 

Van Gogh

In John Berger’s book Ways of Seeing (1972), Van Gogh’s Wheatfield with Crows (1890) is reproduced. The reader is asked to look at the picture, then turn the page. On the next page we are informed that ‘this was the last painting Van Gogh made before he killed himself’ 1 Although Berger resists spelling out what he thinks the conclusions are that the statement invites the reader to draw, one obvious response would be  to see the marks in the painting as expressive, tortured, thence to link them to the artist’s corresponding mental state, specifically, to retrospectively imbue the work with a particular pathos.

Thus, an artwork is reduced in at least two respects. Firstly, to its function as a direct expression of the artist’s mental state, whatever that means, given one is invited to see paint marks as being in some way synonymous with an expressive of a mental state 2. Secondly, as Berger asserts:  ‘It is hard to define exactly how the words have changed the image but undoubtedly they have. The image now illustrates the sentence.‘ (Berger, ibid, p 28, my emphasis)

These reductive moves are made at the expense of a formal analysis in terms of how the brush marks might be understood for the way they fracture or rhythmically consolidate the picture plane in tension with the representational space, aspects that the viewer would need to work out for her/himself, insofar as an explanation of such wouldn’t necessarily make this evident to the viewer. However what concerns me here is not the expressionist fallacy per se, but the process by which extraneous information that has nothing to do with the image in itself, nor is even contextually relevant, is used, in effect, to prop up a work –the work comes to be judged insufficient on its own, and on its own terms: the viewer must be guided to the desired reading. In terms of painting and Van Gogh’s status within its history, it’s the innovations in the handling of paint, the rendering of space, movement, rhythm and colour relations that should really matter.

We are all familiar with the trend for explanatory wall texts in galleries that aim to nudge the viewer into a way of understanding a work, if not explicitly explaining it for them. However, this practice, though pre-emptive, has never been strictly necessary, as long as it has been possible for the viewer to understand a work, assuming a familiarity with its broader context and the language with which it is working. In the examples discussed here, however, contingent information, extraneous to the work and not in any way in it or part of it, is required not only to understand it, but to feel an impact that is wholly dependent on that information to stimulate, if not force, a specific response by providing a story about a pathetic or tragic event to which the work is itself a response.

 

Ai Weiwei

In Ai Wei Wei’s sculpture: Straight (2008-12), hundreds of straightened-out steel reinforcing rods – ‘rebar’- from a collapsed school building are laid in piles on the floor in a ground-hugging, undulating form3. The immediate echo might be of Carl Andre’s works, though Straight is contrastingly curvy and asymmetric. On the surrounding walls are the names of 5000 schoolchildren who perished when the building was struck in an earthquake. Ai Wei Wei has said that his art and politics can’t be disentangled and that Straight is a memorial to the dead children. However, if it is indeed a memorial, can it be a work of art, given the instrumental, didactic function of memorials, regardless of their aesthetic merits? If the work is to be understood as art, then it must surely first be evaluated according to sculptural criteria, via a critical engagement on the part of the viewer: how does it stand aesthetically, as a more or less generic minimalist floor-piece -an arrangement of multiple units of found industrial materials- compared to earlier and current examples? If this is a fair way to approach the work, where does that leave the children’s names on the wall? Are they part of the work in its non-monumental sense? If so, then how are the two related, but if not, what is their purpose?

The issue that concerns me here is the way contingent information is necessary in order to grasp the significance of the work, to feel its impact, indeed, insofar as the origins of the rebar and the circumstances that prompted the work’s creation overwhelm it, one is implicitly discouraged from seeing it in formal terms as sculpture. In this sense there’s an implication that to see it principally as aesthetic is to fail to respond appropriately, so that one’s response may be judged to be less than human. In the case of Straight, it is presumably the poignancy, the melancholic aura, with which the work is imbued, in the knowledge that the rebar was salvaged from the school, disentangled from the young bodies buried with it: empathy trumps aesthetic evaluation. But why, then, have the bars been straightened out? If the work is intended to confront the viewer with the hard reality of the rebar’s origins, why not leave it, forensically, in the state in which it was found?4 Perhaps, the straightening-out functions as a metaphor for the Chinese authorities’ attempts to ‘straighten out’ the aftermath of the event by, for example, refusing to release the names of the dead children. But such a metaphor adds little in any respect, since the response was well documented, and in the same way, the metaphor is also reductive, if all it is intended to do is function as a one-line critique of the Chinese authorities’ response to the tragedy. Art is not a good way of doing politics, but if there is a politics in / of art, it lies in its refusal to function as illustration or in service to pre-existing non-art ideas. Rather it is the resistance art creates to easy understanding, and hence assimilation, that marks its potential political function: we are encouraged to struggle with thoughts and ideas and to raise those struggles to a political principle.

 

Willis Thompson

Another more recent example is the films of Luke Willis Thompson, shown as part of his Turner Prize 2018 exhibit; Autoportrait (2017), Cemetery of Uniforms, and Liveries (2016). Autoportrait shows the face of Diamond Reynolds, who broadcast live on Facebook the moments after her boyfriend, Philando Castile, was fatally shot by police in Minnesota in 2016. Cemetery of Uniforms shows Brandon, the grandson of Cherry Groce, who was shot by police in her Brixton home in 1985, and Liveries shows Graeme, the son of Joy Gardner, who died in police custody in 1993. However, none of this information is in the work.

The three silent movies were shot on black and white 35mm film and closely resemble the Screen Tests Andy Warhol made between 1963 and ’66 (Warhol, whose influence Willis Thompson has acknowledged, used 16mm). In other words, the images contain very little movement and are silent, lasting around three to four minutes each. Insofar as they are simply portraits of what would otherwise be anonymous figures, there’s no way of knowing these people’s life stories except by reading about them. Clearly, the artist intends us to look at these films with the foreknowledge acquired from the explanatory panels at the entrance to the projection space, indeed this knowledge is essential to his intentions for the work, though it is presented as supplementary, in the form of text.

These texts resemble the familiar explanatory panels seen in galleries, except that here the information provides not general context or suggested interpretations, but  specific information about the film’s subjects, without which the work wouldn’t function at all as the artist intended, nor therefore have its desired impact on the viewer. The effect of this strategy is to remove agency from the viewer by interpreting the work for them, thereby denying them the hermeneutic activity that is  fundamental to the experiencing of artworks, while simultaneously reducing them to an  illustrative adjunct. Affect is reduced to being the transfer of information from speaker to listener, in conformity with classic information theory5. Unlike in Berger’s example, however, the films here don’t even illustrate anything, for what would be illustrated? At best one could argue that it’s possible somehow to read-off anguish from the faces of the subjects, but how would one know one was not simply imputing a correlation between facial expressions and inner feelings under the influence of the fore-knowledge acquired at the entrance?  Nevertheless, as with the example of Ai Weiwei, the viewer is similarly directed to respond in a specific way, with the implication that not to do so would suggest a pathological inability to empathise on their part. This coercive effect, and the denial of the viewer’s agency,  is mirrored in the way Willis-Thompson speaks for his subjects, as J J Charlesworth has pointed out: ‘ventriloquising the histories of others to make a political statement goes against the agency of those for whom one attempts to speak.6

In contrast, any knowledge of the sitters in Warhol’s Screen Tests, while informative, is inessential, because the films are about the influence of the shooting conditions on the sitter: the way they negotiate their relationship with the camera, as opposed to their biographies. And in so far as the viewer is confronted with the baldness of the screen test films, we are also implicated in the conditions of the work. The backstories presented with Thompson’s subjects sidesteps the taut relationship between film and viewer that Warhol conjures, but then again without the backstory there’s nothing to it but a Warhol aesthetic plus massive, irrelevant loopers.

We are invited to invest the image with a specific significance that it doesn’t otherwise have. But the image hasn’t changed: it’s the same with or without the information about its subject, therefore the invitation is to read into it something that is not evidently there, in the process of which we surrender ourselves to the manipulations of the artist, our responses over-ridden by the knowledge of the circumstances of the subjects’ lives. What exactly is it that the artist is trying to elicit, and how does that relate to the images? It is as if the films are reduced to illustrations of the contingent facts about them, except that they don’t illustrate anything: they are simply portraits of people. Concomitantly, our response is reduced to an emotional reaction to the circumstances of the subjects’ lives: what else is left, after all, for us to take from the work?

It could be that the work is about photographic meaning, or rather its notorious lack thereof, though this wouldn’t be an original contribution to that debate. One could argue that these films, their context and status as art aside, are no different in themselves from the ‘middle brow art’ practised by amateurs and analysed by Pierre Bourdieu in his book on the subject7 If it were so, why has the artist chosen exclusively black subjects to work with? Perhaps he wants to show solidarity with oppressed minorities, but this fact, combined with the knowledge about them, overrides the viewer’s response, all but eliminating their ability to respond in their own way to the work. In this sense the work isn’t art, it’s a form of very weak documentary, but the artist is too coy to call a spade a spade, to identify himself as really a campaigner, since this could have consequences for his career. In a broader sense there is a problem with artists making careers out of political posturing: posturing because if they were honest they would abandon their art career and re-invent themselves as activists, doing real political work in an arena where it might have some traction. The irony is that in reducing the desired response to the films in this way, the artist is denying his own work any potential complexity or subtlety it might have: he’s shooting himself in the foot. As if conscious of the post-colonialist issues raised by the white artist wanting to ‘help’ his black subjects, Willis-Thompson met and discussed the project with Diamond Reynolds in order to secure her co-operation, but insofar as the project remains his work and property, this merely pushes the problem back without resolving it.

 

Jaar

In Alfredo Jaar’s photograph The Eyes of Gutete Emerita (1996), a photo of a woman’s eyes, there is an attempt to get the viewer to put themselves in the place of the subject and to imagine, assuming it’s even possible, the experience the woman actually lived through, of witnessing the hacking to death of her husband and two sons during the Rwanda massacres. In this respect the work offers a challenge on the power and limits of the imagination, connecting it to the debates around the unrepresentability of such occurrences, paradigmatically, the Holocaust. Nevertheless, in the sense in which it seems to suggest that trauma or any other kind of emotional state can be read off the image, it is problematic. As an image it is simply a photograph of eyes, and, as with Willis-Thompson’s films, we are asked somehow to see something in the image that isn’t really there. The suggestion seems to be that the witnessing of the events is somehow recorded in the image of the eyes and legible as such, but how can this be? Alternatively, it could be seen as a critique of the limits of representability, but this seems unlikely: why would an artist choose a specific, highly traumatic event to photograph to mount a conceptual work about the limits and possibiities of meaning in  photographs?The distinguishing feature of art is that the viewer must struggle to make sense of the work as it is. The more he/she is guided to a pre-determined response, the less like art it is. Why do artists feel they have to illustrate or even respond to political events by making artworks about them? While there is of course a venerable history of this, we don’t for example, appreciate Goya’s 3rd of May 1808 mainly for what it depicts, but for his skill as a painter in handling a traumatic subject. While the work is charged by the subject it doesn’t stand or fall on account of it and, from the point of view of this essay, explanatory texts are not required to provide information that is not in the work and hence not a part of it. Of course, it would be equally reductive to talk only about the aesthetic value of Goya’s paintings, or Turner’s The Slave Ship (1840), his dramatic painting of sick slaves being thrown overboard to their deaths, but the difference between the these works and the ones discussed above is that the subject is inscribed in the work, so although it could be said to be illustrative on one level, there is an abundant wealth and complexity in the way subject and handling are intertwined.

 

1.John Berger: Ways of Seeing, Penguin Books, 1972, pp 27-28. (While one of his last, it almost certainly isn’t the last painting he made).

2.See Hal Foster: ‘The Expressive Fallacy’, in Recodings, Art, Spectacle, Cultural Politics, Seattle, Washington: Bay Press, 1985, for a critique of the idea that expressionist art offers a uniquely unmediated access to the artist’s intentions.

3. Straight was shown at the Royal Academy in 2015 as part of a survey of Wei Wei’s work.

4. I am grateful to Amy Dickson for this observation.

5. See Claude Shannon: A Mathematical Theory of Communication, Bell System Technical Journal, 1948, pp 379-423.

6. J J Charlesworth, ‘Turner Prize 2018’, Art Review, Jan-Feb 2019.

7. Pierre Bourdieu, 1996. Photography, A Middle Brow Art, Polity press.

 

 

Grainy Realism: Four films by Vicky Smith.

Short essay on four films by Vicky Smith for a forthcoming DVD.

The four films on this disc exhibit continuities, in single screen format, with a preceding phase of work, undertaken for a PhD project, which began with rotoscoping figures by scratching into clear celluloid, to which body fluids and physical matter were added in subsequent films, to performances in which an inked bicycle tyre is ridden along a filmstrip, which is then projected.

The performativity of these precedents is abundantly present in much of the work here, from the precise handicraft of Small Things Moving in Unison (2019) to the exuberant physicality of Noisy Licking Dribbling and Spitting (2014). In the latter a regularly pulsing, livid red splat, filled with bubbles, envelopes the screen, gradually increasing in pace. We hear background radio noises and optical sound generated by the image, and as the pace increases to a climax there is a piercing frequency burst, generated by a printed section of optical soundtrack that then appears briefly in the picture area. At this point the pace slows for a while, yellow and orange fields displacing the saturated red of the first half of the film. Finally, the image breaks up into continuous snaking forms over a field of speckles and the sound grows in volume and intensity.

Noisy Licking measures out its pace in tongue-sized stretches, and, in its use of body parts to generate the image complements the earlier 33 frames per foot (2013), in which the artist inked her bare foot and imprinted it onto the film strip. Both films may be seen as rude counter-blasts to the altogether softer and more garishly colourful work of Jennifer West, who often works with cosmetics, such as lipstick kissed onto the film in Rainbow Party (2008). The use of contingent bodily measures; tongue, foot (Smith’s foot measure 33 frames of 16mm, against the 40 frames of the imperial, 12 inch measure) interspersed with blank pauses, calls to mind Stan Brakhage’s Mothlight (1965), in which rhythm is determined by the length of the various materials used until the end, where images are spaced momentarily along the film, like punctuation marks.

Noisy 2

In Small Things Moving in Unison, the simple, classical, but technically exacting method of scratching black film to create white marks is given a vibrant, virtuoso treatment. The film begins with a field of dancing dots that seem to grow and diminish in size individually, as well as coalescing into different compound shapes relative to the framing edges; rectangular blocks reform into narrow, horizontal spirals, thence into discrete balls, sometimes with hexagonal patterning. Two distinct kinetic processes are evident; the movement of the dots relative to each other, and the sections between frames of image that are left black in order to create an overall on-off rhythmic pattern. Frequently, the centre of the screen is empty, defined by a border of dots which regularly spread out until they hit the limits of the frame, at which moment a sharp, straight line appears, a gestalt formed by the eye-brain as it seizes on a regular, familiar shape. This happens frequently, establishing a play on shapes contained by the frame, and shapes that spread beyond it, but in the latter moments the framing edges become part of the image, holding it in, so that we don’t have a sense of off-screen in the usual way when something moves out of frame.

Small Things 2

Both Noisy Licking and Small Things are impressive in the way scale, kinesis, pace and rhythm are controlled and developed. Towards the end of Small Things the white dots become larger and resemble tears in the film, as if actually ripped out of the film strip instead of scratched into it. The implied sounds of Noisy Licking are deployed literally in Small Things, through Shirley Pegna’s soundtrack, which consists of busy, wordless vocalising; whispers, tongue-clicks and taps, sucking, moaning and mutterings, which grow to a crescendo as the white dots give way to the torn shapes, as if the shapes are caused by the sonic activity. Just as the soundtrack of Noisy Listening is briefly disturbed by the burst of synthetic tone, so in Small Things a knot of fine hair appears momentarily within / behind one of the larger holes, breaking the heterogeneity of means. One wonders how it got there. At the same time the holes are colour-fringed, suggesting the film was made from a strip of exposed colour film, and not black and white as the film has appeared hitherto.

 

Primal is in some respects a pared down companion to Small Things, beginning as it does with single, feint scratched shapes that flash up, interspersed with black pauses. The shapes grow larger and appear to have been bleached to produce gradations of desaturated colour. Dense nets of scribbled lines appear, as if cancelling the shapes, which themselves threaten to burst and escape the confines of the frame. Where Small Things is delicate and dancing, Primal is closer to Noisy Licking in mood. The film is considerably enhanced by Shirley Pegna’s complex, music track, a combination of concrète and instrumental sounds which, in its very scratchy, insistent quality, avoids falling into the trap of mood-creation. Both film and sound have room to breathe, so that one doesn’t feel one’s attention is divided, as it so often is.

Primal

Not  (A)Part nods to Mothlight but also to Mothfight (1985), a much less well known film by Vanda Carter, a fellow filmmaker of Smith’s at the London Filmmakers’ Co-op. However, whereas Mothlight’s material lies flat on the film plane, the images constituting Not a Part are three-dimensional, photograms of ‘53 dead bees found on walks in southwest Britain’, as is stated in the text that appears near its beginning. This clear reference to the disastrous decline in the bee population, and indeed the insect population generally, is addressed in the shapes of dead bees, which are shown in negative, reanimated by their movement through the projector. The film serves as a kind of quasi-scientific study, a gathering of data, but put to different ends. In terms of themes and ideas this film differs from the others yet is of course connected in its being a camera-less animation.

Not A Part

This long-established way of working has a rich history, which Smith has added to and enriched further with these four films, which range from the exquisite, finely wrought movements of Small Things Moving in Unison to the grimy, abject slapstick of Noisy Licking Dribbling and Spitting.

4 x 3 x 2: New multiple film loop work

New 16mm film loop work for two to four projectors. 4 x 3 x 2 was shot on Super 16mm, thereby generating an optical soundtrack when played on a standard 16mm projector. It was premiered at BEEF -Bristol Experimental and Expanded Film- as part of a one-day workshop with Vicky Smith, BEEF member, on Rayogramme and site specific film making on October 23rd, 2018. The version shown here is for two projectors.

Interval [ ] still : now is on at Tintype Gallery, London, during March 2018, with work by Moyra Derby, Jost Münster, Conor Kelly, Joan Key and myself. Thanks to Teresa Grimes of Tintype.

Images and video link below. All the work is designed to overlap in various ways, and further layers are developed by the mobility of the spectator. My 16mm film loop Standard 8, is a site specific work, in that the subject of the film is projected back onto itself in negative. The silhouette of the painting by Moyra Derby is in the film, while the actual work is adjacent to the projection. The second work Black White Black, is a two projector digital video, designed to be adaptable to different situations, here projected onto a free standing work by Moyra Derby and paintings by Jost Münster.

Nicky Tintype New Work

Wider view below shows a hanging work by Jost Münster.

P1020507Interval [ ] still : now