Aspects of Stereoscopy

Countless stories have been handed down to us over the years about artists vying with each other to see who was more successful at playing tricks on the senses. Let us take just one well known example from the many anecdotes:1 the artist Parrhasios observed some sparrows flying in to peck at a bunch of grapes depicted so realistically in a mural painted by his fellow artist Zeuxis. Later he invited Zeuxis into his atelier to show him that he, too, was a master of this art. Once inside Parrhasios’s studio, Zeuxis reached out to draw aside the curtain that was apparently concealing the other’s picture, only to realize that what he thought was a curtain was no curtain at all, but indeed the picture itself. While the skill of one artist had succeeded in tricking the birds, the skill of the other was capable of tricking even a fellow professional. The wealth of evidence, anecdotal or otherwise, testifying to the deceptive effects of pictures on animals and human beings is undeniable; trompe l’oeil curtains achieved great popularity in 17th century art and even in our own times the photorealistic painting of the 70s was a response to the public demand for this kind of optical subtlety considered the highest form of artistic mastery. This style has always sought to emulate nature even in its finest detail and has striven to overcome the distinction between illusion and reality.

This tradition reached its climax in the period of the Renaissance when the development of the concept of linear perspective first made it possible to produce the three-dimensional expansion of a body in space correctly proportioned on a flat picture surface. In the year 1425 Filippo Brunelleschi conducted the following experiment:2 in the central doorway of the cathedral in Florence he positioned his painted panel showing a perspective view of the Baptistery, which stands on the opposite side of the piazza. The vanishing point of his picture lay precisely in this central doorway, but Brunelleschi positioned his panel with the unpainted side facing the observer. A small peephole in the middle of the panel was all that permitted the observer to see through it and across the piazza to the Baptistery; but then, when a mirror was held up — blocking his line of vision — the observer was able to see Brunelleschi’s identical painting on the front side of the panel. If contemporary accounts are to be believed, the impression must have been quite stunning, and this effect was further enhanced by the use of an additional mirror surrounding the picture of the building which reflected the clouds in the sky over the real Baptistery onto the first mirror. This was a breathtaking demonstration of the effects to be achieved by this new method of presentation: henceforth, the visible world was subjugated to the mechanical principles of rational construction and could be reproduced exactly.

Yet Leonardo himself complained of the deficiencies of even the best pictures of this kind. His criticism focused on their effort to present spatial illusion in comparison to the plasticity of real objects. He recognised that this qualitative difference in our perception had to do with our visual capacity for binocular vision as opposed to the mere monocular perspective of a painting.3 Nevertheless, it took more than three centuries until the highly talented physics professor Charles Wheatstone finally discovered the principles of stereoscopy. When he was reporting on the results of his research in the summer of 1838, he did not miss the opportunity to point out the duties facing the next generation of artists who now, thanks to his discoveries, had been put in a position “to draw and paint the two component pictures, so as to present to the mind of the observer, in the resultant perception, perfect identity with the object represented”; he was looking forward, he said, to stereoscopically depicted “flowers, crystals, busts, vases,” and the like, which could “thus be represented so as not to be distinguished by sight from the real objects themselves.”4

Whether the objects mentioned represented Sir Charles’s personal taste in art is not something we want to argue about here; but if he particularly selected them because, as motifs for pictures, such banal subjects would not be so distracting for the viewer (who is often preoccupied with interpretation and superfluous associations), he was indeed anticipating one of the essential aspects of stereoscopy. In any case, this would be consistent with his own preference for line drawings reduced to simple, clear geometrical figures so that he could carry out his experimental research free from any distractions. In a similar way, any artistically motivated concept of stereoscopy will exclude from the outset all the questions which seek — often in vain — to define the presumed intentions of the creator and which in fact do more to obstruct than to promote the acceptance of his work: the less an observer has to puzzle over what the artist wanted to say, the more he is able to concentrate on the medium, which itself is the one part of the message.

The other part delves into the profound complex of non-art problems which arise from the sensational discovery of stereoscopy and which lead to metaphysical considerations; these in turn have striking consequences with regard to the (supposed) illusionist character of the image of space on the one hand, and of the reality content of the space of the image. Let us briefly explore a line of thought which may bring us somewhat nearer to the mysteries of stereoscopy.

As the story of Zeuxis and Parrhasios illustrated, an attempt to deceive the senses does not necessarily need to be part of the original wish to comply with the rigours of correct perspective. It is clearly possible to achieve illusionist effects without perspective construction and, similarly, perspective effects can be achieved without stereoscopic duplication. Accordingly, stereoscopy does not necessarily have anything to do with this kind of art derived from linear perspective, which blurs the distinctions between illusion and reality, as was so masterfully demonstrated by Brunelleschi, even if it initially appears to be using the same artistic means. As a result of their orientation towards one single focal point, conventional paintings — and photographs — can, at best, have the effect of optical illusions: for no matter how well they work, their effect generally does not endure. By contrast, the stereoscopic spatial image is triggered by the merging of two pictures whose different perspectives correspond to the parallax of the distance between the eyes. While each of the eyes records the corresponding partial picture (viewed through a stereoscopic device) as optically correct, the spatial image is first perceived when the observers mind performs the necessary fusion; the deception — if indeed there is one — does not take place in the eyes of the observer but in his brain, and even though he is aware of it, it will remain; it is inescapable. The sensory imprints on the retinae are (almost) identical, whether they come from normal vision or from looking through a stereoscopic device; in both cases the sensory processes are the same. The spontaneous impression of three-dimensionality results from the unavoidable fusion of the two retinal images generated by real-world objects — or by the stereoscopic picture pair as substitutes — and in both cases produces an authentic perception.5 Unless we wish to proclaim that our entire world is nothing more than a phantasmagoria and hallucination, we must accept the reality of the stereoscopic effect; the visual space is real and — both in normal vision and in the stereoscope — it is no surrogate, nor is it a fata morgana.

This authentic perception is not merely a possibility, indeed it must be a reality for us because space — although it repeatedly appears as a realontological structure — is not an object of perception but rather a way of perceiving objects. According to Immanuel Kant’s critical analysis of the principles of knowledge, space only can “be the form of all appearances … under which alone external appearance is possible”6 i. e. “a necessary a priori representation which underlies all outer appearances”,7 but not a quality attached to the things themselves. The phenomenon of stereoscopy is only understandable under this premise because the spontaneous sensation of space invoked by the two two-dimensional pictures cannot be achieved empirically; in fact there is no three-dimensionality here, and there is actually no reason why it should be seen at all. Thus we find ourselves in the middle of a most significant aporia, which — as the successor to Kant, and with explicit reference to the invention of the stereoscope — was pointed out by Arthur Schopenhauer. Thus, given that binocular fusion is indeed physiologically located in the cerebral system, space is the product of an organ for which it, itself, must always be a precondition.

The point at which the metaphysics of perception turn so spectacularly into the metaphysics of space and vice versa is nowhere better illustrated than in the simulation of visual perception by stereoscopy — and precisely because of its restriction to two parallactically different perspectives which create three-dimensionality where, in fact, there is nothing more than two-dimensionality. In this sense, a hologram, with its profusion of multi-perspective plasticity, has become too common to evoke any astonishment. By contrast, looking into a stereoscope never ceases to be a mystery — and a quasi-dialectic mystery, too — proceeding from thesis (left picture), to antithesis (right picture) and synthesis (fused picture), an entirely anti-materialistic dialectic, and not progressing in separate steps but immediately and simultaneously: as the viewer looks, the duplicity of the stereoscopic twins is instantly neutralized [referring to Hegel’s triple word-meaning8] in the identity of their imaginary unification. While each one gives the other what it is lacking, they each lose — albeit only temporarily — their own individuality and textural consistency as they provide the material substrate for this virtual “third”, which they — by disappearing — create, but which does not exist for real.

At the same time, Charles Wheatstone’s discovery was the conception of a completely new type of imagery without historical precedent and quite unique; and one whose hidden artistic potential has, as yet, hardly been explored at all. The fundamental paradox of stereoscopy resulting from the smooth interplay between the unreal within reality and the reality of the unreal is what generates — in addition to the speculative aspect mentioned — not least, its particular aesthetic attraction. And this is ultimately what characterizes stereoscopic art — a kind of artistic hermaphrodite — as a singularity within the fine arts: between and yet beyond painting and sculpture. Thus, as a logical and ultimate consequence of the classic traditions of painting, this singularity reveals for all to see the genuine ambivalences of vision and visibility, to which it owes its very existence.

© 1990 (2007) Achim Bahr, Aspekte der Stereoskopie. — Article in the vision RAUM exhibition catalog, Neckar­werke, Fellbach 1991; published under the same title in 1 + 1 = 3, Museum for Art and Art History, Goch 1993; revised by the author, translated by David Gough 2007. — This text is an abridged version of another article partially published by the author, Stereoscopy as Applied Metaphysics, cf. idem. On Stereoscopic Painting, catalog for the exhibition Stereoskopie, Berlin Technical Museum (formerly: Museum für Verkehr und Technik, today: Deutsches Technikmuseum Berlin), Berlin 1989

  1. See Ernst Kris / Otto Kurz: Legend, Myth, and Magic in the Image of the Artist: A Historical Experiment; preface by Ernst Gombrich, Yale University Press, September 1981 ↩︎
  2. Brunelleschi is generally held to be the inventor of linear perspective; the historical basis for this can be found in eye-witness reports by Antonio Manetti, his friend and admirer, and Brunelleschis biographer. Incidentally, the experiment was successfully reenacted by students of Florence University; cf. Joscijka Gabriele Abels: Erkenntnis der Bilder (Realization of the Pictures), Teil I, Abs. 3, Frankfurt/New York 1985, 79ff, but also Paul Kurt Feyerabend: Wissenschaft als Kunst (Science as Art), Chapter 1, Frankfurt 1984, 17ff ↩︎
  3. Cf. Leonardo da Vinci: Trattato della pittura; Treatise on Painting (Codex urbinas latinus 1270), trans. Amos Philip McMahon. Princeton, N. J.: Princeton University Press, 1956, Part two, Section II F, No. 197 and particularly Part three, II. main section, Fascicle No. 4 ↩︎
  4. Charles Wheatstone: Contributions to the Physiology of Vision, Part the First, on some remarkable, and hitherto unobserved, phenomena of binocular vision, §4, in: Brewster and Wheatstone on Vision, edited by Nicholas J. Wade, London, New York e. a. 1983, 72ff ↩︎
  5. This is why Sartre’s strict distinction between imagination and perception does not appear appropriate here either; cf. Jean-Paul Sartre: The Imaginary: A Pheno­­meno­logical Psychology of the Imagination, Routledge Chapman & Hall, 2004 ↩︎
  6. Immanuel Kant: Critique of Pure Reason, §3b ↩︎
  7. Ibid., §2, paragraph 2, loc. cit. ↩︎
  8. Translators note: in the original German text the word “aufgehoben” was used which has – in Hegels sense – the meanings of superseding, cancelling out, and neutralizing. The most appropriate equivalent to be found was the adjective insubstantial with its meanings of non-material, not physically strong, of no significance, unreal or difficult to discern, but this, admittedly, does not work as well as the German original. ↩︎