Stereoscopic Media


This article was updated on 10 Oct 2013, and is filed under Research Provocations.

Ryan Pierson Research Provocation: A New Haptical Cinema

lens flare into darkness

Because scholarship has focused so intently on the addition of depth in 3D, there exists the widespread assumption that the 3D image demands a more embodied response by the spectator. Separating the image into different focal planes makes the objects onscreen seem to share our space.  This, we tend to assume, elicits a stronger bodily reaction than a flat screen.  Such an assumption undergirds the frequent comparisons of 3D to the shock-based cinema of attractions and Thomas Elsaesser’s claim that 3D is catching up to the vibrations of surround sound (2013).  But our bodies count on more than depth.  For a thing to appear touchable, it must have volume. Depth is a relation of objects to each other in space; volume is a palpable sense that an object itself occupies space.  3D projection has depth but no volume.

This cardboard-cutout effect of stereoscopy is well-known.  Jonathan Crary writes: “[T]he fundamental organization of the stereoscopic image is planar.  We perceive individual images as flat, cutout forms arrayed either nearer or further from us.… Compared to the strange insubstantiality of objects and figures…the absolutely airless space surrounding them has a disturbing palpability.” (1990) The effect is especially visible in subtitles (as recently seen in Pacific Rim (2013)) and lens flare (as seen recently in Star Trek into Darkness (2013)).  In 3D projection, these elements occupy the foremost plane of vision while remaining flat.

The distinction between depth and volume has important consequences for how we talk about space in 3D projection.  For German art historian Alois Riegl, depth demanded a disembodied form of viewing; volume evoked the sense of touch.  Riegl famously argued that ancient Egyptian art (especially in bas-relief) engendered a haptic sense of space.  Renaissance painting, with its geometric perspective and plays of light, created optical space.  He argued that “with an increased space and three-dimensionality the figure in a work of art is also increasingly dematerialized” (1983, quoted in Lant 1995).

2D cinema relies on roughly the same cues of light and perspective as Renaissance painting, and on these grounds we may still want to assert 3D as more haptic.  But Riegl’s distinction is based on whether or not a figure is actionable: whether it can be touched, walked around, examined from different angles.  3D cinema fulfills none of these criteria.  The surface of the screen is no longer a barrier to the film’s world – yet, paradoxically, this makes the things in the film more strongly felt as surfaces.  How might we describe this sensory confusion?

Two accounts of cinematic space are helpful in this regard.  First:  Noël Burch’s theory of space in early cinema reverses Riegl’s haptic/optical terms (1990).  Instead of distance or emptiness, Burch’s optical space is flat.  This conception commands early cinema’s use of stage backdrops, even lighting, a perpendicular and unmoving camera, and so on.  It is most visible in the work of Méliès.  Haptic space emerged with classical film style in the 1910s, suggested through varied camera angles, dramatic lighting, and camera movement.  Haptic space is deep and traversable; we can imagine stepping into it.  Optical space is fixed and confrontational.  (These oppositions are played with in Hugo (2011) when clips from various Méliès films get 3D treatment – see my “Méliès in Stereopsis” (

Thomas Lamarre’s theory of space in cel animation makes a slightly more complex distinction (2009).  Lamarre calls the ability to move in depth across a homogenous field “cinematism” Analogous to Burch’s haptic space, cinematism is tied to a desire to conquer space.  “Animetism,” by contrast, occurs when characters and backgrounds are drawn on separate celluloid sheets.  These sheets can create depth by moving in relation to each other.  Animetism is not flat: movement and depth are created with parallax, moving different layers of the image at different speeds.  But this movement of layers yields a special kind of depth that calls attention to the gaps between planes.  We can move laterally across these spaces, but not into them.  Animetic space seems to exist independently of us, heterogeneous and uncontrollable.

These accounts still assume the primacy of the screen, and so cannot be directly applied to 3D.  But they are suggestive models for thinking about the strange ways our bodies are simultaneously hailed into and banished from 3D space.  We engage its surfaces as well as its depth.



Burch, Noël.  (1990) Life to Those Shadows.  Trans. Ben Brewster.  London: BFI Publishing.

Crary, Jonathan.  (1990) Techniques of the Observer: On Vision and Modernity in the Nineteenth Century.  Cambridge: MIT Press.

Elsaesser, Thomas. (2013)  “The ‘Return’ of 3-D: On Some of the Logics and Genealogies of the Image in the Twenty-First Century.”  Critical Inquiry 39.2, 217-246.

Lamarre, Thomas. (2009) The Anime Machine: A Media Theory of Animation.  Minneapolis: University of Minnesota Press.

Pierson, Ryan. (2012)  “Méliès in Stereopsis.”  The Funambulist,

Riegl, Alois.  Late Roman Art Industry (1901).  Trans. Rolf Winkes.  Rome: Giorgio Bretschneider Editore, 1983.  Quoted in Antonia Lant, “Haptical Cinema.”  October 74 (Fall 1995), 45-73.

One Comment

  1. Esther JACOPIN
    February 20, 2015

    Dear Sir,

    I carefully read your provocation. There are a few points on which I would like to have your opinion.
    As you write in your provocation, the audience cannot touch, walk around, examine from different angles stereoscopic images projected on the screen like you can do with sculptures.
    Claude Bailblé — French professor at Paris university (Paris 8 – Saint-Denis), believed that we shouldn’t say “3-D”.
    Indeed, stereoscopic cinema still works with an audience sitting in front of a screen. If you move your head while watching a stereoscopic movie, it will have no effect on the image. Your movement will not change your “point of view” on the image, as it does in immersive stereoscopic systems (like the Cave Automatic Virtual Environment also known as CAVE or, more recently, the Oculus Rift). That’s the reason why Claude Bailblé asserts that we should say “2D and a half” instead of “3-D”.

    But I wish you studied examples from S3D films that try to pass trough this difficulty. I am thinking of Wim Wender’s stereoscopic 3D (S3D) films (“Pina” and “If buildings could talk”), in which the 3D rig, mounted on a steadicam or a crane, is constantly moving, even in still shots. These slight movements permit to change the perspective, to film from different angles the environment in order to enhance the depth and the volume in the stereoscopic image.

    Because stereoscopic images do have volume. We all know that stereoscopy reproduces the human binocular vision. Thanks to the inter-ocular distance between the two cameras that reproduces the parallax phenomenon. Due to physics and optic laws, stereoscopic images do have volume. I think it is a little confusing when you assert that “3D projection has depth but no volume”. What difference do you make between S3D images and “projected” S3D images?

    It is true, however, that lens flare and subtitles have a “disturbing palpability”, “while remaining flat”. But I am surprised that you consider examples naturally flat from converted S3D films (for which the stereoscopy was the result of a financial decision instead of an artistic choice). Who has ever said that light, a physical phenomenon, is three dimensional? Subtitles are only bi-dimensional layers over stereoscopic images…

    When I started reading your research provocation, I thought you intended to propose a new analysis of stereoscopic cinema through Riegl’s and Burch’s thesis. You conclude by suggesting that there is an opportunity to rethink the audience physical relation towards S3D space. Gilles Deleuze, when analysing painter Francis Bacon’s work, employes the term “haptic” for pictorial phenomenons in which the eye “reads” an image like the hand does when touching an object. Considering the stereoscopic cinema three spaces (behind-the-screen space, screen space and out-of-the-screen space), what is the balance between the (whole) body and the (only) eye in this “New Haptical Cinema”?

    Yours sincerely,

    — Esther JACOPIN

Leave a Reply

− 2 = seven