Painting and visual reference in Kubrick

A shot can recall a painting without there being any copying involved. When studying pictorial references, it is useful to distinguish what we observe in the images from what the documents and participants allow us to claim about their preparation.

A visual comparison does not, by itself, identify a source

The resemblance between two images may lie in the arrangement of bodies, the direction of the light or the relationship between colours. These similarities allow us to describe an affinity. They do not prove that a film crew used a particular painting, since the same compositional problems can be solved in similar ways within broad visual traditions.

A useful comparison should indicate which elements coincide and which do not. The viewpoint, the scale of the figures and the distribution of space matter just as much as a recognisable motif. Saying that a scene looks like a painting is not enough unless we specify what relationship we can observe. Resemblance should become a description, not a badge of prestige.

The organisation of symmetry and perspective offers tools for that description. A painting and a shot may share an axis yet create different experiences. In cinema, a character can leave their position, a voice can divert our attention, and a cut can change the meaning of what we have just seen. The image is not separate from the sequence’s duration.

The relationship with paintings that appear within a set requires a different distinction. A painting hanging on a wall is a visible object within the fiction. It may help characterise a space without having served as the model for its composition. Confusing these two functions turns the material presence of a work into a claim of influence that still needs proof.

A testimony can show how a reference is transformed

In an interview published in 1976, John Alcott told American Cinematographer that, while preparing Barry Lyndon, they studied the lighting effects of the Dutch masters and opted for lighting that came more from the side. His statement documents a reference and a decision to transform it. The material they studied is not presented as an instruction to be copied without change.

This case helps us understand visual research as an active process of comparison. A crew may study an image to identify a quality and decide that it needs to be altered. The reference helps frame a problem: how much volume the faces should have, how the bodies should stand out from the background, or what relationship there is between brightness and depth. The cinematic solution responds to the particular conditions of filming and performance.

The analysis of composition and zooms in Barry Lyndon explores the specific scenes. Here, what matters is the type of evidence: Alcott speaks of preparatory work and a lighting choice. His testimony does not allow us to assign the name of a particular painting automatically to every shot, nor does it imply that all the film’s images come from the same pictorial tradition.

Collaboration must also remain visible. A reference may be discussed by the director, cinematography, production design and costume departments, each with different problems to solve. A painting brings together relationships on a single surface that filming has to distribute across space, bodies, materials and light. Translating a visual impression requires the work of several crafts, not simply recognising a reproduction in a book.

Duration also transforms what a painting holds simultaneously. A figure may move from shadow into light, or a composition may be revealed gradually. Cinema has to decide when we see each relationship. A static reference can guide a result that unfolds over time without that result becoming an unmoving reproduction.

Research materials need context

The research for Napoleon brought images together with historical information and location materials. This collection makes it possible to study what kinds of representation were available during development. It does not mean that every reproduction preserved in the archive was chosen for a scene. A preparation archive may contain alternatives, rejected options and materials consulted for very different questions.

A costume image may be of interest for a detail of its cut; a portrait, for how it conveys status; a depiction of a battle, for the arrangement of troops. These functions need to be established through context, annotations or testimony when we want to attribute a specific use. Its mere presence in a folder does not explain all the decisions it may have prompted.

The use of photography as visual research raises similar problems. Paintings and photographs may coexist in a dossier without providing the same kind of information. An image made in another period may document a representation of the past, rather than the past directly. Its date and purpose matter when it is used to support a reconstruction.

We must also distinguish a comparison proposed after a film’s release from a reference used before filming. The former may be analytically fruitful even if it does not explain the production process. The latter requires historical evidence. Neither needs to displace the other: they can answer different questions about the film and how it came to be made.

A measured formulation might say that a shot shares a visual organisation with a painting, that a document preserves a reproduction, or that a participant acknowledged studying a particular tradition. Each statement attributes a different degree of relationship. Keeping them separate makes it possible to analyse Kubrick’s visual culture in detail without treating resemblance as proof or reducing artistic research to a search for copies.

Production source: Herb A. Lightman, ‘Photographing Stanley Kubrick’s Barry Lyndon’, interview with John Alcott, American Cinematographer, March 1976. Research documentation: La Cinémathèque française, Stanley Kubrick virtual exhibition, ‘Napoleon: Method’; University of the Arts London, Stanley Kubrick Archive.

View all articles in Artistic and literary reference points for Kubrick