Pictorial Ambiguity. Approaching ›applied cognitive aesthetics‹ from a Philosophical Point of View
Autor: Nicolas Romanacci
[erschienen in: IMAGE 10 (Ausgabe Juli 2009)]
Schlagwörter: pictorial ambiguity, applied cognitive aesthetics, epistemology, media art, Wittgenstein, Duck Rabbit, Giotto, Morandi, Cézanne, Nicolas de Stael, Jasper Johns, Dan Graham
Disziplinen: philosophy, epistemology, image science
1. Introduction: Wittgenstein, Gombrich and the ›duck-rabbit-picture‹. Seeing as thinking – ambiguity and art. Introducing my major concerns.
»Ambiguity – duck or rabbit – is clearly the key to the whole problem of image reading.« (Gombrich 2002: 198)
Figure 1: The Duck-Rabbit (›H-E-Kopf‹), Ludwig Wittgenstein: Philosophische Untersuchungen (1958). Frankfurt/M. [Suhrkamp] 1975, p. 309
In his Philosophical Investigations (Wittgenstein 1975) Wittgenstein takes the ambiguous duck-rabbit-picture (fig. 1), (Wittgenstein 1975: 309) to discuss the usage of the term ›seeing‹. That leads him to his conception of »aspect seeing«, described (in an initial attempt) as »half visual experience, half a thinking« (Wittgenstein 1975: 314). This statement can be related to Wittgenstein’s view that »›Seeing as ...‹ does not belong to the act of perception« (Wittgenstein 1975: 313), because it is always an act of interpretation and understanding, therefore thinking. Wittgenstein keeps questioning the nature of the relation between seeing and thinking, and later on he changes his initial conception of an addition of these two acts (or: of the act of ›noticing an aspect‹) and rejects his own formula »Seeing+Thinking« to state that the answer should rather be that »many of our concepts do cross here« (Wittgenstein 1975: 339). It is impossible to untangle the relation between the visual and the mental, there’s only a misleading separation in theoretical discourse (see also: Schürmann 2008: 128 et sqq.). Wittgenstein reveals this grave theoretical misconception and turns the »duck-rabbit« into a paradigmatic case. It is quite astonishing how Wittgenstein develops his intricate investigation on the nature of (the usage of the term) seeing by starting with and refering to such a simple – and by the way quite humorous – drawing like the duck-rabbit-picture.
The same picture (in another version, see Kihlstrom 2006 for a history of this figure) is used by E.H. Gomrich in his Art and Illusion (Gombrich 2002: 4) to introduce »his major concerns: the nature of vision and of representation, and the problem of reconciling the objectivity of the latter with the conventionality and the relativity of vision.« (Goodman 1972: 141)
Like Wittgenstein, his ›path of investigation‹ leads him to the insight that »we have come to realize more and more, [...] that we can never neatly seperate what we see from what we know.« (Gombrich 2002: 331) Comparing Wittgenstein’s and Gombrich’s views, related to the »duck-rabbit« and the phenomenon of ambiguity, William G. Lycan extracts that
»The ›bucket theory of the mind‹ is false. The mind is not a passive receptacle in which sense-data are ›deposited‹, ›processed‹, or ›habitually correlated‹; it is rather an active organ whereby pre-established concepts and perceptions interact. The having of ›sense-data‹ is not epistemologically prior to the perception of physical objects or to seeing a painting ›as‹ something – if anything, the priority is the other way round.« (Lycan 1971: 230)
While Wittgenstein and Gombrich seem to »complement each other quite well« (Lycan 1971: 236) and though they offered some absorbing insight into the phenomenons of »seeing as« and ambiguity and about their role in aesthetic appreciation, there remain some questions open (to pose new ones was quite an intent of Gombrich, see: Gombrich 2002: 332) and my major concerns can be introduced:
How can the phenomenon of ambiguity be described more exhaustively and critically from a philosophical point of view, and how can theoretical differentiation sensibilise for a differentiated and concrete aesthetic analysis of artworks. Furthermore, how can these questions be related to each other and to the general status of ambiguity in consideration of the cognitive conception of ›seeing as a form of understanding‹. Summarized I want to introduce my project on ambiguity as an attempt of ›differentiation on three levels‹: a theoretical, a media, and a general level. I understand my investigation also to be interpreted as exemplifying an approach to an ›applied cognitive aesthetics‹. The following article is an excerpt of my study, providing some insight into the ›media level‹. I will conclude with some considerations on ambiguity from an epistemological point of view.
2. ›Media Level‹: Steps towards a typology of pictorial ambiguity as a base for aesthetic analysis. Differentiating ›figure-ground‹, ›figuration-abstraction‹, ›spatial‹, ›narrative-temporal‹, ›semantic‹, and ›categorial‹ ambiguities. From Giotto via Morandi to Jasper Johns. With an extension to an exemplary analysis of Media Art, regarding two works of Dan Graham.
Challenge at hand is to develop a typology of pictorial ambiguity, meant as an instrument and a point of departure for an aesthetic analysis.
On a theoretical level we can treat the duck-rabbit-picture (fig. 1) and the Necker cube (fig. 2) as similar cases of »multiple meaning« (see Scheffler 1979). But for an aesthetic analysis it makes sense to differentiate here and discriminate between two types of pictorial ambiguity. In a general sense, one could say, that both pictures might be labelled as cases of ›semantic‹ ambiguity, for the meaning at hand is ambiguous. But more specifically the Necker cube case could be labelled as a case of a ›spatial‹ ambiguity, because the indecision regards strictly speaking a spatial interpretation. Third, one could consider another quite classical figure, the Wittgensteinian Doppelkreuz (›double cross‹), illustrating another important type of compositional problem, the ›figure-ground‹ ambiguity.
Figure 2: Louis Albert Necker: Necker Cube, 1832, http://en.wikipedia.org/wiki/Necker_cube, 15.08.2009 see article on ambiguity and epistemology
Figure 3: Double cross (›Doppelkreuz‹), Ludwig Wittgenstein: Philosophische Untersuchungen (1958). Frankfurt/M. [Suhrkamp] 1975, p. 331
2.1 ›Figure-Ground-Ambiguity‹: Combined with an ambiguity between figuration and abstraction. An exemplary study correlating works of Morandi, Cézanne, Nicolas de Staël and James Abbott McNeill Whistler.
As introduced, a typology of pictorial ambiguity is not merely meant as an aim of inquiry, but rather as one possible basic instrument for subsequent aesthetic analysis. What has to be pursued is analytical method of subtle differentiation, combined with comparative elements. In the following I would like to propose some short exemplary case studies, following this procedural method.
Beyond an analysis of compositional structure, it is particularly intended to gain some insight into the artist’s compositional strategies, presuming that they serve as extremely important clues for understanding an artwork. This approach is also based on an utterance of Max Imdahl in his Giotto-study, where he is refering to the conviction, that every fair aesthetic analysis has to extract from »the work as a problem-solution« the antecedent artistic problem, for this is the real final aim of a science of art (Imdahl 1996: 75).
Following this general approach, I would like to introduce my first case of »figure-ground« ambiguity with a work from Giorgio Morandi, that gives way to retrace the artist’s decision to focus and forge a compositional strategy.
The following drawing (fig. 4) is easily detectable as a typical Morandian still life arrangement. Though already quite fairly abstracted, the depicted subject is still observable as a group of six objects: four bottles, an object behind the third bottle from left and a bowl in front of them, standing on some plain, indicated through a horizontal line. The degree of abstraction might be evaluated as rather easy to decipher, as this kind of an outline drawing would count as a familiar and comprehensible way of depiction.
Figure 4: Natura morta, 17 x 24 cm, Giorgio Morandi: Catalogo generale, disegni. Milano [electa] 1994, p. 146, fig. 7
As mentioned already, the degree of abstraction might count here as quite unproblematic and no ›figure-ground‹ ambiguity is in sight. But already some small rearrangements of the compositional network change the situation in an amazingly radical and drastic way for the artist’s focus has switched fundamentally to an accentuation of a ›figure-ground‹ ambiguity and a heavy intensification of abstraction (fig. 5).
Figure 5: Natura morta, 16,5 x 24 cm, Giorgio Morandi: Catalogo generale, disegni. Milano [electa] 1994, p. 146, fig. 6
To confirm my assumption, that this rearrangement has a bearing on a compositional strategy, let’s take a look at a series of water colour, Morandi worked on, in 1959. Obviously he took his drawing as a model to be transfered to another technique (fig. 6).
Figure 6: Natura morta, 21 x 31 cm, Giorgio Morandi: Catalogo generale, acquerelli. Milano [electa] 1991, p. 129, fig. 40
Fairly similiar compositional structures are to be observed at that time in a work series of at least 17 water colours. It seems that this problem was of great interest to Morandi. He switched from some cases with still detectable figurative parts to an increasing abstraction, and also a cumulative accentuation of the ›ground‹ elements. In the following example Morandi seems to have reached a transgression of a balanced proportion between figure and ground elements in favour of highlighting the ground elements, with a strong effect on the abstract character of the whole work. In contrast to the preceding example, he accentuated the second element from left through using a different colour (blue) than for the other parts (fig. 7).
Figure 7: Natura morta, 16,2 x 20,7 cm, Giorgio Morandi: Catalogo generale, acquerelli. Milano [electa] 1991, p. 129, fig. 39
If an increasing accentuation of abstraction and of the ›ground‹ elements serves as a compositional strategy, the following two works might count as an advanced stage of executing the latter. For this interpretation would count that the following works are from the late work of Morandi (fig. 8 and fig. 9, both 1962).
Figure 8: Natura morta, 16,5 x 24 cm, Giorgio Morandi: Catalogo generale, disegni. Milano [electa] 1994, p. 226, fig. 66
Figure 9: Natura morta, 114,1 x 19 cm, Giorgio Morandi: Catalogo generale, acquerelli. Milano [electa] 1991, p. 180, fig. 5
Compare John Clark’s view on Morandi’s strategy:
»[A]mbiguity of forms occurs in Morandi’s modest still-lifes and landscapes when volumes and spaces subtly interlock. Background space penetrates the foreground objects fusing them into a solid alloverness of surface [...] Ambiguity in Morandi, the interaction between objects and their shadows or between figure and ground, is a visualization of transformation itself. It is transformation in action. This interaction between figure and ground is not a diagrammatic demonstration of trick spatial structure, like an Escher drawing or an illustrative Dali mindscape. [W]e experience an organic living surface as the painting’ makes itself before the viewer’s eyes. The searching process of transforming these simple bottles and vases into a picture lives on in the viewer’s presence.« (Clark 1983)
Though here’s not enough room to deliver an exhaustive analysis, including historical aspects, it makes sense for our purpose, to take a look at some more examples. I would like to focus attention hereby especially at considering what compositional strategy each of the following artists presumably and considerably pursued.
In the late works of Cézanne, we find several cases, where it is hard to decide, if the work at hand is just ›unfinished‹, or if the artist’s will to »realisation« has forced him to radicalize his prescinding from figurative restraints (see the catalogue: Vollendet – Unvollendet 2000). Even if we can’t decide, if the works at hand (fig. 10 and fig. 11) would have counted for Cézanne as nearly finished or rather as an abortive attempt, in comparison especially with the last example from Morandi, we may not be too far of considering a comparable artistic strategy.
Figure 10: Maisons sur la colline, 1904-06, 59,7 x 79,3 cm, Paul Cézanne: Vollendet – Unvollendet (catalogue). Ostfildern-Ruit [Hatje Cantz] 2000, p. 347, fig. 122
Figure 11: Le Jardin des Lauves, ca. 1906. 65,4 x 80,9 cm, Paul Cézanne: Vollendet – Unvollendet (catalogue). Ostfildern-Ruit [Hatje Cantz] 2000, p. 147, fig. 147
Another artist, who radicalized the problem of expressing the tension between figuration and abstraction, combined with a well balanced ›figure-ground‹ strategy, is Nicolas de Staël. He forces the ambiguous impression through entitling his works with quite concrete titles. See: Figur am Strand, and Bildnis Anne (fig. 12 and fig. 13).
Figure 12: Nicolas De Staël: Figur am Meeresstrand, 1952, 161,5 x 129,5 cm, Ruhrberg et. al.: Kunst des 20. Jahrhunderts. Band 1. Köln [Taschen] 1998, p. 219
Figure 13: Nicolas De Staël: Bildnis Anne, 1953, 130 x 90 cm, Ruhrberg et. al.: Kunst des 20. Jahrhunderts. Band 1. Köln [Taschen] 1998, p. 231
Nicolas de Staël’s struggle with the tensions between the figurative and the abstract gives way to understand a comparable struggle from an artist, who’s work at the first sight might be far of from abstract tendencies. James Abbott McNeill Whistler entitled his famous portrait of his mother as an Arrangement in Grey and Black (fig. 14). It is interesting, that regarding the tactics of entitling the work, here we will find a case somehow just the other way round. Whistler always heavily insisted that his work has to be interpreted predominantly as an abstract composition (»arrangement«) and not as an (mere) illustration or representation (see: Gombrich 1997: 530)
Figure 14: James Abbott McNeill Whistler: Arrangement in Grau und Schwarz, 1871, 144 x 162 cm, E. H. Gombrich: Die Geschichte der Kunst. Frankfurt/M. [Fischer] 1997, p. 531
2.1.1 Representation, Exemplification and Expression. Applying Goodman’s terminology.
»[S]ometimes the subject serves merely as a vehicle for what is exemplified or expressed.« (Goodman 1978: 106)
I would like to provide some elaboration on Goodman’s terminolgy, regarding the notions of denotation, exemplification and metaphorical exemplification. Whistler’s Arrangement in Grey and Black serves perfectly to re-read some crucial passages from Goodman’s Languages of Art, taken from the chapter on exemplification and expression, The sound of pictures. The following passage can be even transfered directly to an observation of Whistler’s painting at hand:
»Before me is a picture of trees and cliffs by the sea [in our case: a picture of an old woman], painted in dull grays, and expressing great sadness. This description gives information of three kinds, saying something about (1) what things the picture represents, (2) what properties it possesses, and (3) what feelings it expresses. The logical nature of the underlaying relationships in the first two cases is plain: the picture denotes a certain scene and is a concrete instance of certain shades of gray. But what is the logical character of the relationship the picture bears to what it is said to express?« (Goodman 1976: 50)
Goodman’s ingenious solution to analyse the phenomenon of expression is to interprete the latter as ›metaphorical exemplification‹. Whereas ›exemplification‹ can be described as ›possession plus reference‹, expressing that not all of an artwork’s features have referential function for its appearance as a symbol. Like in the most cases the weight of an artwork does not matter for an interpretation. For our investigation it seems important to me to stress Goodman’s insight that for an aesthetic analysis exemplificational reference is in the majority of cases much more crucial for understanding an artworks meaning than its denotational ›content‹. Whistler’s insistence that the importance of his composition is to be seen much more in the »arrangement« of colors than in the representation of a person serves as a perfect example for being aware that »sometimes the subject serves merely as a vehicle for what is exemplified or expressed«. On the other hand the notion of exemplification opens ways to handle ambiguous interpretations of non-figurative and abstract works. I will come back to that point later on, discussing a work of Jasper Johns.
2.2 ›Spatial‹, ›Narrative-Temporal‹ and ›Semantic‹ Ambiguity: Giotto as an exemplary case.
Having started with Morandi’s way of »figure-ground« ambiguity, we crossed some fairly abstract works from Nicolas de Staël to finally detect strategies of ambiguity in a comparable figurative work from 1871. Also in earlier periods we can analyse strategies of ambiguity. For this chapter I want to give a short idea of the amazing works of Giotto, starting with occurances of ›spatial‹ ambiguity.
2.2.1 ›Spatial‹ Ambiguity: Giotto as an exemplary case (part 1).
To introduce my investigations on Giotto, I would like to refer to a study from – interestingly no art historian, but – a philosopher, John H. Brown. In his study Unscrambling Giotto’s Space (Brown 2004) he investigates occurances of spatial ambiguity in Giotto’s work. He visualises Giotto’s strategy through digital manipulation. Brown takes Giotto’s work Expulsion of Joachim from the temple (fig. 15) to illustrate the occurances of some spatial inconsistencies. He lists 4 points, while particularly in this work point 3 is not observable:
Figure 15: Giotto: Expulsion of Joachim from the temple, 1304-06. 200 x 185 cm, http://www.wga.hu/index1.html, 15.08.2009
To envision Giotto’s spatial construction, Brown shows firstly a classical diagramm:
Figure 15, illustration (Brown, J. H): http://www.philosophy.umd.edu/Faculty/jhbrown/Giotto/index.htm, 15.08.2009
According to the construction lines it is apparent that: »The temple enclosure is presented in essentially parallel projection; the other diagonals converge, but to horizons on different elevations; and the pulpit is far too small for the indicated distance.« (Brown 2004)
To make Giotto’s characteristic spatial strategy even more clear, Brown develops some digital manipulated images of the original scene. See again the original (fig. 15) and two manipulated images (fig. 15, mod 01 and fig. 15, mod 02).
Figure 15, digitally manipulated 01: http://www.philosophy.umd.edu/Faculty/jhbrown/Giotto/index.htm, 15.08.2009
Figure 15, digitally manipulated 02: http://www.philosophy.umd.edu/Faculty/jhbrown/Giotto/index.htm, 15.08.2009
For my purpose it isn’t necessary to discuss Brown’s considerations how to evaluate this phenomenon, I just want to list some of his questions and suggestions:
»Comparing the two we can ask how exactly the regularity of the transform affects us? What artistic good in the original is enhanced or degraded? Does the irregularity of the projection in the original take anything from it that Giotto or his contemporaries would have valued? Does it take anything away from what we value in Giotto?
The loss or gain, since aesthetic, must be expressed in terms of aesthetic properties. For instance it might be alleged that the scrambled projection makes for a more coherent surface design.« (Brown 2004)
To get a step closer to an answer, I want to discuss another example, The Marriage at Cana (fig. 16). Though at the first sight the spatial ambiguity in this picture seems to be a very subtle and therefore seemingly secondarily, Max Imdahl’s brilliant analysis (Imdahl 1996: 75 et sqq.) offers an exemplary reading to understand why Giotto might have chosen to pursue spatial ambiguity as a compositional strategy.
Figure 16: Giotto: Marriage at Cana, 1304-06, 200 x 185 cm, http://www.wga.hu/index1.html, 15.08.2009
Figure 16, illustration: Giotto: Marriage at Cana, 1304-06, 200 x 185 cm, http://www.wga.hu/index1.html, 15.08.2009
For our purpose, we don’t have to discuss the whole analysis of Imdahl, though I recommend a reading warmly. It serves us sufficient to summarize, that the horizontal line (see highlighted at fig. 16, ill.), correlates perfectly to the line we can draw starting from Jesus and aiming at the person drinking the wine. As this line correlates to the central subject of the image – Jesus converting water into wine – it seems accurate to detect Giotto’s reasons for offering a »wrong« spatial construction as his solution of the underlaying artistic problem: to illustrate and express the unsettling wonder through a unsettling spatial ambiguity.
Evaluated at a general level, we might describe the process of detecting such a ambiguous composition as a source of aesthetic pleasure for the viewer, like Marc de Mey did in his article Mastering Ambiguity.
»The horizontal line [...] approaches the status of an ambiguous figure. Though it seems to be a single homogenous entity, only the middle segment consists of a section parallel to the picture plane, while the outer segments consist of sections orthogonal to the picture plane. While it has a major twodimensional compositional function as a single geometrical element, in three dimensions, the curtain line arises from a more complicated composition. One could obviously conjecture that this is a coincidence resulting from the height of the viewer‘s vantage point. But even if the painter chose the vantage point on those grounds, there remains a mild tension evoked by this ambiguity. The disambiguation of it, [?] without any doubt largely unconscious, leading to unpretentious aesthetic pleasure.« (De Mey 2006: 273)
I would like to critizise his view of a need to ›disambiguate‹ the scene. Recent studies in cognitive psychology serve to support my critique:
»We do not believe that ambiguity in art needs a complete resolution. It might be an art-inherent feature that a residual ambiguity might be left open and accepted by the perceiver. This is likely because otherwise it would have to be assumed that there is only one correct solution to the challenge of art. This is not the case. Artworks can often be experienced aesthetically several times, yielding different solutions like, for instance, when the artwork is perceived again with more expertise.« (Leder et al. 2004)
»Ambiguities of the Third Dimension« can be observed in various manifestations. In his Art and Illusion Gombrich gives some analysis on works like Escher woodcuts (Gombrich 2002: 206) via a watercolour from Klee (Gombrich 2002: 223) or ancient Mosaics (Gombrich 2002: 226). Compared with all these examples, with Gombrich cubism could be evaluated as »the most radical attempt to stamp out ambiguity«. As an exemplary case serves a work of Braque, The Table from 1928 (fig. 17).
Figure 17: Georges Braque: Still Life: The Table, 1928, http://www.superstock.com/stock-photos-images/998-118, 15.08.2009
Beyond aiming at a final evaluation, it seems to be a conclusive assumption to understand ambiguity as a crucial precondition and clue to open the possibility for various interpretive considerations. Let it be to explain cubism »as an extreme attempt in compensation for the shortcomings of one-eyed vision« or the observation of its succeeding »in countering the transforming effects of an illusionist reading« (Gombrich 2002: 239).
2.2.2 ›Temporal-Narrative‹ Ambiguity: Giotto as an exemplary case (part 2).
The following example serves as a special kind of ambiguity. I propose to call it ›temporal-narrative‹. Again we can’t discuss the composition in detail, and again I refer to the study of Imdahl. Example (fig. 18) shows Giotto’s interpretation of the Raising of Lazarus. The underlaying artistic problem of this scene could be described as following: How can the subsequent narrative scenes of Jesus calling »Lazarus, Come Out!« (›t3‹), Lazarus doing so (›t4‹); the precending lamentation of the women (›t1‹) and the opening of the tombstone (›t2‹) be adequatelly depicted in one pictorial scene. Giotto offers an amazing solution. If we look closely, the 4 correlative temporal areas can be detected as illustrated in (fig. 18, ill.).
Figure 18: Giotto: Raising of Lazarus, 1304-06, 200 x 185 cm, http://www.wga.hu/index1.html, 15.08.2009
Figure 18, illustration: Giotto: Raising of Lazarus, 1304-06, 200 x 185 cm, http://www.wga.hu/index1.html, 15.08.2009
The crucial point is the way how Giotto arranged these areas to enable the seemingly impossible depiction of subsequent scenes in one pictorial space. If we look closely at the women, it is obvious, that they do not look at Jesus. They rather seem to look out of the scene. That serves as experiencing them like somehow on another pictorial level as the scene behind them, subsequently we could interpret them as appearing at time level ›t1‹. The two persons holding the tombstone are correlated to the women through kind of a mirrowed composition. The right person is even looking directly at the viewer, we could place them at time level ›t2‹. Even more difficult seems the problem of depicting Jesus’s utterance and Lazarus’s subsequent ›coming out‹ in a simultaneous scene. Giotto did find an absorbing solution in relating the subsequent scenes through the person (with green clothes) right in the middle between them. The person’s body is somehow twisted, his right hand reached out to Jesus while looking at Lazarus. These subtl compositional strategies seem to enable the impossible: depicting subsequent scenes in one single image, we could call this point an entanglement of the time levels 3 and 4. To emphasise Giotto’s unique genius in doing so, let’s compare his solution with the work of an unknown artist, depicting obviously the very same scene (fig. 19), very probably refering to Giotto’s earlier version. In (fig. 19, ill.) it is illustrated, that the composition unlike Giotto’s is not able to express a ›temporal-narrative‹ ambiguity. For the described 4 narrative scenes are rather correlated to each other like happening at the same moment in time. Because unlike in Giotto’s composition, the women are looking directly at Jesus, so does the person at the right side. And the crucial function of the person in the middle between Jesus and Lazarus is missing in this version. In disambiguating the scene this artist seems to have also lost crucial aesthetic quality (compare Imdahl 1996: 73).
Figure 19: Unknown artist: Raising of Lazarus, after 1306, http://www.wga.hu/index1.html, 15.08.2009
Figure 19, illustration: Unknown artist: Raising of Lazarus, after 1306, http://www.wga.hu/index1.html, 15.08.2009
2.2.3 ›Semantic‹ Ambiguity: Giotto as an exemplary case (part 3).
Closing my Giotto series, I would like to give two examples for what I want to call ›semantic‹ ambiguity. In both cases, a diagonal line serves to express an ambiguous interpretation. In the Annunciation (fig. 20) the diagonal of the hill can be interpreted as supporting Jesus’s decline, but at the same time his following resurrection.
Figure 20: Giotto: Lamentation (The mourning of Christ), 1304-06, 200 x 185 cm, http://www.wga.hu/index1.html, 15.08.2009
Figure 20, illustration: Giotto: Lamentation (The mourning of Christ), 1304-06, 200 x 185 cm, http://www.wga.hu/index1.html, 15.08.2009
In the infamous Arrest of Christ (fig. 21), the virtual diagonal line expresses at the one hand Jesus’s inferiority through being arrested, but at the same time – following his predominant glance – his absolute indomitability. For an enhanced interpretation see again Imdahl 1996.
Figure 21: Giotto: The arrest of Christ (Kiss of Judas), 1304-06, 200 x 185 cm, http://www.wga.hu/index1.html, 15.08.2009
Figure 21, illustration: Giotto: The arrest of Christ (Kiss of Judas), 1304-06, 200 x 185 cm, http://www.wga.hu/index1.html, 15.08.2009
126.96.36.199 ›Semantic‹ Ambiguity: The need for a dedection of compositional correlations.
I want to stress that I hold it basically necessary to correlate a presumed ›semantic‹ ambiguity to compositional occurances. Looking at the infamous Mona Lisa (fig. 22), said to be ambiguous for her miraculous smile, we can clearly correlate this semantic ambiguity to the technique of ›sfumato‹, keeping the angles of her mouth in an unclear shadow.
Figure 22: Leonardo Da Vinci: Mona Lisa (La Gioconda), ca. 1503-05, 77 x 53 cm, http://www.wga.hu/index1.html, 15.08.2009
The following example, Vermeer’s The Girl with the Pearl Earring (fig. 23) might be also described as ambiguous, like Semir Zeki did in an absorbing way:
Figure 23: Johannes Vermeer: Girl with a Pearl Earring, ca. 1665, 46,5 x 40 cm, http://www.wga.hu/index1.html, 15.08.2009
»[A] higher level of ambiguity is to be found in the multiple narrative interpretations that can be given, for example, in Vermeer’s painting entitled The Pearl Earring. Note that this is a single stable image, and the only variable is that the brain of the beholder can offer several equally valid interpretations of the expression on her face. She is at once inviting, yet distant, erotically charged but chaste, resentful and yet pleased. These interpretations must all involve experience and memory, of what a face that is expressing these sentiments would look like. The genious of Vermeer is not that he provides an answer but, by a brilliant subtlety, manages to convey all the expressions, [...] Because there is no correct solution, the work of art itself becomes a problem that engages the mind.« (Zeki 2006: 262)
Though offering a convincing general evaluation of ambiguity, and though Zeki’s view might be comprehensible to the most of us, in this case I would consider it difficult to correlate the expressed ambiguity to specific compositional features.
2.3 ›Categorial‹ Ambiguity: Jasper Johns as exemplifying philosophical problems.
»These paintings illustrate a philosophical problem, but they do it to envision an aesthetic problem, that we are not invited to solve, but to experience.« (Richard Wollheim 1982: 210)
Jasper Johns Flags (fig. 24) are broadly interpreted to be ambiguous, as the following exemplary statements illustrate:
Figure 24: Jasper Johns: Flag, 1954-55, 107,3 x 153,8 cm, http://www.moma.org/modernteachers/large_image.php?id=196, 15.08.2009
»The critic Robert Rosenblum asked of Flag: ›Is it blasphemous or respectful, simple-minded or recondite?‹ Its brilliance, implied his review, lay in these questions.« (Jones 2003)
»Johns recognized the artistic potential of Duchamp’s Readymades, but probably of greater importance, he was fascinated by the elaborate, sly games Duchamp played with incompatible aesthetic ideas – with the definition of art and non-art; ›real‹ objects and ›art‹ objects; with the connection of verbal and visual images, of the optical and tactile, of the literal and the illusionistic, of the two- and three-dimensional, of what is conceptualized and what is seen, that is, with the complex and ambiguous process of experiencing art.« (Sandler 1978)
My account of a need to correlate an asserted ambiguity to traceable compositional strategies can be achieved in this case through an analysis of John’s treatment of the pictorial surface. Through providing a heavily structured plain, highlighting the uniqueness of every little irregularity, and therfore the uniqueness of the painting at hand, Johns stresses an ambiguous relation to the – in general – more universal character of the depicted flag as a symbol. Transfered to Goodmanian terminology we have a clear case of an ambiguous relation between the pictures denotation (as a »flag-symbol« for the idea of a state) and what the picture exemplifies (through its structured surface), respectivily expresses (through metaphorical exemplification). (for a discussion of Johns regarding Goodman see also: Hölscher 2005 ) Analysed in this way, there’s also no problem to interpret Richard Wollheims utterance of Johns work as ›illustrating a philosophical problem‹, as clearly doing this through metaphorical exemplification.
»In the case of Johns it is surely not only crucial that it is problematic, but also that it has been also intended to be so, and the ambiguity is one we are manouvered into. If there‘s a way out of this ambiguity then not because of an error of Johns. These paintings illustrate a philosophical problem, but they do it to envision an aesthetic problem, that we are not invited to solve, but to experience.« (Wollheim 1982: 210)
If Wollheim understands John’s work as illustrating a philosophical problem, it might also serve as an interesting case to relate it to the idea of investigating an artworks features regarding the conceptions of nominalism or platonism. For a platonist would rather claim that basically and finally the idea of a state is crucial to be expressed by an artwork, a nominalist would be more interested in taking a closer look at the concrete features of the object itself and its forms of reference as a symbol. For Johns’s ›flag‹ is able to express both ways of interpretation, it is its ambiguity that holds responsible for enabling these complex considerations and thus it is its ambiguity that originates the artwork’s ›power and richness as a symbol‹.
2.4 Beyond the Image. ›Perceptual Ambiguities‹ in Media Art: An exemplary study on the work of Dan Graham, regarding questions of time, space, identity and category.
»I think art is about ambiguity [...].« (Graham 2009a)
»Media Art« is characterized through an extension of pure pictorial forms of expression. We can distinguish categorial labels like »installation«, »performance«, »cybernetical modells«, »feedback«, »immediacy«, »liveness«, »interface«, »interaction« (Kacunko 2004), »virtual«, »telematic«, »digital« or »genetic« (Grau 2003 and 2007). In the following I want to extend my analysis of pictorial ambiguity to ›Media Art‹. I have chosen the work of Dan Graham as an exemplary case for several reasons. Graham worked in a variety of mediums, like text, photography, performance and installation (cc-video and his ›pavillons‹) (see Brouwer 2002 and Hamilton 2009 for a comprehensive overview of Graham’s work) and his work can be related to many labels listed above, like ›installation‹, ›performance‹, ›feedback‹, ›»immediacy‹, ›liveness‹, even ›interface‹ and ›interaction‹, if understood in a broad sense. Graham’s work could be characterized in the most general way as exemplifying modes of ›multiple and complex reference‹, as Graham always aims at a critical, ironical and sometimes humorous distance to simplified conceptions, like the often propagated ›neutrality‹ of ›Minimal Art‹. I want to show that Graham’s strategies to create ›multiple and complex reference‹ can be often related to occurances of ambiguity. I want to elaborate this phenomenon by analysing two works: Public Space/Two Audiences (1976) and Time Delay Room 1 (1974).
2.4.1 Dan Graham’s Public Space/Two Audiences (1976): Questions of space, identity and category.
Dan Graham’s Public Space/Two Audiences was realized the first time as part of the 1976 Venice Biennale. The theme of the 1976 Biennale was the environment, with reference to architecture. At the Biennale usually artists show their works in one of the several seperate exhibition spaces, the so called »pavilions«. These pavilions are partly organized in thematic exhibitions, such as 1976 Ambiente, organized by the Biennale itself (for context and history of the work see Francis, in Hamilton 2009: 182 et sqq.).
Dan Graham decided not to exhibit ›a work‹ in the traditional way of showing an object in a space. He used the space itself to be structured in the following way (fig. 25 and fig. 26).
Figure 25: Dan Graham: Public Space/Two Audiences, illustration, 1976, Marianne Brouwer: (Ed.): Dan Graham. Werke 1965-2000. Düsseldorf [Richter] 2002, p. 173
Figure 26: Dan Graham: Public Space/Two Audiences, installation view, 1976, Marianne Brouwer: (Ed.): Dan Graham. Werke 1965-2000. Düsseldorf [Richter] 2002, p. 173
The structure of the work causes some complex effects, regarding questions of the categorial status of the work as an (art-) object, related at the same time and inextricably to questions of status and identity of the (seemingly) distant viewer as a subject watching an artistic production. Public Space/Two Audiences causes a reversal: Instead of watching an artistic production (created for the art market), enclosed in a spatial environment, the viewers are »exhibited« themselves, caused through the materiality and structure of the »box«. At the same time – in opposition to the supposed »neutrality« of the material construction – the social and psychological aspect of the pavilion becomes evident. This aspect stands particularly against a possible first view on the used materials (thermopane, glass, mirrow) as exemplifying the stereotype of the »neutrality« of »Minimal Art«, and the propagated possibility of a perceptual immediacy. But perception takes always place in time, and on the (inevitable) second view, and while watched together with other visitors over a specific time span, there emerges an increasing contradiction: the material and structure of the room are experienced as control instances of psychological and social behaviour (Graham 1976, in Brouwer 2002: 172 et sqq.). The visitor gets involved in an ambiguous realm of questions related to his own identity: is he a viewer of an object (the work), or is he already a part of the work, or more radically speaking, is he and the other viewers somehow the work itself? Is he an observer of other visitors or is he part of the observations of the others? Is he acting as an individual or as part of a group? These questions are fundamentally supported by the multiple layered spatial ambiguities caused through the mixture of glass, mirror (only on one side of the room), and the reflections in the glass. The seemingly neutral materiality enables ambiguous consciousness processes.
»[T]he observer is made to become psychologically self-conscious, conscious of himself as a body which is a perceiving subject; just as, socially, he is made to become aware of himself in relation to his group. This is the reversal of the usual loss of ›self‹ when a spectator looks at the conventional artwork. There, the ›self‹ is mentally projected into (and identified with) the subject of the artwork. In this traditional, contemplative mode the observing subject not only loses awareness of his ›self‹, but also consciousness of being part of a present, palpable, and specific social group, located in a specific time and social reality and occurring only within the architectural frame where the work is presented. In Public Space/Two Audiences the work looks back [...].« (Graham 1976, in Alberro 1999: 158)
Dan Graham emphasizes the fundamental importance and inevitability of temporal aspects of experiencing a work. Especially regarding the consciousness process of the viewer. In Public Space/Two Audiences temporality and questions of consciousness have been important aspects of the work. In the Time Delay Room series from 1974, relations between temporality and consciousness are the central theme. In the following part I want to show how in Time Delay Room 1 these central aspects are related to ambiguous structures.
2.4.2 Dan Graham’s Time Delay Room 1 (1974): Questions of time and consciousness.
The basic spatial structure of Time Delay Room 1 is comparable to the one of Public Space/Two Audiences. Again we find two seperate spaces with the same measurements. But unlike the former, the two rooms are connected through an opening and a complex closed-circuit-video-system is installed (fig. 27 and fig. 28).
Figure 27: Dan Graham: Time Delay Room 1, installation plan, 1974, Marianne Brouwer: (Ed.): Dan Graham. Werke 1965-2000. Düsseldorf [Richter] 2002, p. 149
Figure 28: Dan Graham: Time Delay Room 1, illustration, 1974, Marianne Brouwer: (Ed.): Dan Graham. Werke 1965-2000. Düsseldorf [Richter] 2002, p. 149
The work has been shown 2001 at the ›zkm‹, karlsruhe, as part of the exhibition ›ctrl[space]: rhetorics of surveillance‹. The following descriptive exhibition-text points at the ambiguous character of the work structure:
»Description: two rooms of equal size, connected by an opening at one side, under surveillance by two video cameras positioned at the connecting point between the two rooms. The front inside wall of each features two video screens – within the scope of the surveillance cameras. The monitor which the visitor coming out of the other room spies first shows the live behavior of the people in the respective other room. In both rooms, the second screen shows an image of the behavior of the viewers in the respectively other room - but with an eight second delay.
On entering, you can cast a swift glance at the viewers and the screens in the other room, but you are invited to focus initially on the room where you are. On the screens there you see twice [live, and with a delay] the other viewers. However, on both screens you also see [if smaller, as an image within the image] the two monitors in the other room. On studying them for a while you will notice that on the one there is a live transmission of what is going on in the room where you are, for you can relate this footage directly to the behavior of people in the room where you are. And you can spot on the other monitor in the other room, symmetrically, the same shot as on one of the screens in the room where you are. This still offers no hint as to how the two screen images showing the behavior of the viewers in the respectively other room relate to each other. Both images are contemporaneous and show the other set of viewers from the same angle, but the behavior is inconsistent. This questions the first assumption, namely that both are live transmissions. If a person moves from the other room into the room where you stand then this act takes place on both screens, but with a great difference. The screen showing the live transmission shows the person leaving the room until s/he is no longer covered by the camera and then, for a few seconds, you see the behavior of the people in the other room minus this person, and finally this person has manifestly become part of the group of people in your room. However, the moment the person becomes part of the set of viewers in the room where you are standing you see them on the screen in the other room leaving that room, i.e., their intention to change viewer group. Since the viewers watching the screens have their backs turned to you, there can be no unprejudiced comparison of the different impressions, and the relationship between the two screen images remains ambiguous. (Stemmrich 2001)
The ambiguous appearance of the screen images, combined with the specific time delay of 8 seconds, enables an ambiguous consciousness experience. A time-lag of 8 seconds is said to be the outer limit of the neurophysiological short-term memory (Stemmrich 2001). While watching yourself on the screen (recorded 8 seconds ago), you will therefore not relate the recording to a past state of mind, but to your present consciousness state. As this inconsistency will be always related to the recorded and observed behaviour of the other visitors, there’s also created a feeling of being part of a group of ›observed observers‹. Like in Public Space/Two Audiences a seemingly neutral and ›simple‹ structure of material and media causes a complex and ambiguous consciousness experience.
The current Dan Graham show at the Whitney (see publication: Hamilton 2009) is entitled Dan Graham: Beyond. His work is described as having always been pointing beyond: »beyond the art object, beyond the studio, beyond the medium, beyond the gallery, beyond the self.« I guess that creating ambiguity is a fundamental strategy for Graham to enable such transgressions. Let it be particularly his installations, which structures could be described to function »in the mode of ambiguity« (Garrigues 2009). The exhibition text of the show at the Museum of Contemporary Art, Los Angeles, supports my impression from a more general point of view by stating that Graham »successfully develops a phenomenology of ambiguity« (Bedford 2009). And in another overview of Graham’s work, it is summarized that »of all the themes that Graham explores, ambiguity remains his most consistent« (ArtandCulture 2009).
The works of Dan Graham served as exemplary cases of extending and transfering an analysis of occurances of pictorial ambiguity to an analysis of work structures, that extend the pure pictorial. While the perception of Graham’s work is still comparable to pictorial forms through it’s mainly visual perception, I want to stress, that the basic structure of the appearance of ambiguity can be also transfered from the visual to other realms of perception, like the auditiv. Returning to our initial duck-rabbit example and Wittgenstein’s conception of »seeing as«, I want to finish this part with a consideration of William G. Lycan regarding a comparison of seeing and hearing, giving hints for »powerful generalization«:
»It is interesting to note that we can speak of ›hearing as‹ in the same way. A person can hear a single tone as a tonic, as a fourth degree, or as a leading-tone. And Wittgenstein’s criterion for ›S is hearing the note as‹ applies here: we know that S hears it as a leading tone rather than as some other degree only by how he resolves it when given the chance (and we notate it accordingly as a# instead of a bb). The foregoing remarks about ambiguity, and certainly Gombrich’s points [...] can be powerfully generalized in this regard. Such ambiguity is what makes modulation effective; and in the last century composers have made extensive use of tonal ambiguity per se, leading up to atonality, to deliberately complete and thoroughgoing ambiguity.« (Lycan 1971: 233)
Some additional comments to my short ›typology‹: ambiguity is surely not meant to be the only or most important aspect or way of reference of art in general or a single artwork my 7 ›types‹ do not cover every kind of ambiguity a ›type‹ of ambiguity always needs to be correlated to other specific features, to compositional structures and (historical) contexts of the artwork at hand the titles of my ›types‹ are meant as working-titles. I do hold the focusing on explication to be much more important than insisting in ›the final caption‹ an investigation in the appearances of ambiguity should be rather a starting point for further aesthetic analysis than an aim.
3. Conclusion: Some epistemological considerations. The argument from conflicting truths.
»The notion of the structure of a work is as specious as the notion of the structure of the world.« (Goodman 1972: 127)
In the following part I would like to discuss some epistemological aspects regarding ambiguity. I will start with some quite classical considerations, rejecting the idea of the »objective« or »innocent« eye. This will lead us to the most radical epistemological arguments regarding ambiguity and art, refering to some more Goodmanian conceptions, rejections and ›reconceptions‹.
3.1 Ambiguity, perception and the (rejection of the) notion of ›the given world‹: One argument against naive empiricism in the philosophy of science
In one of the standard works of guides to the philosophy of science, Alan F. Chalmers illustrates right in the beginning of his book What is This Thing Called Science? the shortcomings of naive empiricist accounts of science with a classical ambiguous figure 25, the Schrödersche Treppe (Chalmers 1999: 9, see: Kalkofen 2006 for some interesting remarks on that figure and in general on ambiguity and Gestalt-Psychology).
Figure 29: H. G. Schröder, (Helmholtz, this version): Schrödersche Treppe, 1910, Alan F. Chalmers: Wege der Wissenschaft (1976). Berlin et.al. [Springer] 2007, p. 9
Already the title of Chalmer’s chapter summarizes its central point, that ›visual experiences are not affected by the image on the retina‹ (my translation). We’ve got to reject the naive view of the possibility of an objective ›recording‹ of the world by simply looking at it. Two viewers will not necessarily see the same ›object‹ under considerably equal circumstances. The Schrödersche Treppe serves as an example to clarify this argument. Easily we can follow the description that this figure can be understood as depicting stairs ›seen from above‹, or as ›seen from below‹ (comparable to the Necker-Cube). But at the same time it seems undoubtful that the image(s) on the viewer’s retina(s) didn’t change. Furthermore, our first insisting that the pictures ›subject‹ are at least some stairs is shattered through experiments, showing that some african clans, that are not used to three-dimensional depiction, didn’t see any stairs, but a two-dimensional pattern (Chalmers 1999: 9). We will probably follow the conclusive evaluation easily and willingly, that there are eligible doubts at hand regarding the view that ›facts‹ are given to the viewer directly through the senses, though at the same time it seems undoubtful, that there is only one, fixed world given to the viewers, independend of what they are ›seeing‹. At this seemingly harmless point and despite the seemingly safe ground of the notion of ›the one world given to us‹, we have in fact reached slippery ground, from an epistemological point of view. I don’t want to discuss or critizise Chalmers’s position in the following, but I want to discuss some arguments, making clear, why Chalmers had to insist explicitely (Chalmers 1999: 12) on his adherence on ›a given world‹, independent from any observer: he is aware that the notion of ›the one and only given world‹ is not so undoubtful as it may seem at the first sight.
3.2 ›Many worlds if any‹: The argument from conflicting truths. Related to Goodman’s nominalism, constructivism, relativism and irrealism.
The Schrödersche Treppe served as an example to show that every (sensual) observation is fundamentally dependent on the observer’s interpretation. If we follow that argument, it is clear that perception is no passive act but a genuin active and cognitive procedure. Goodman is absolutely following this approach:
»The mind then is actively engaged in perception just as it is in other modes of cognition. It imposes order on, as much as it discerns order in a domain.« (Goodman 1988: 6)
But Goodman fundamentally disagrees with the notion of »the one and only independet world« we refer to in the end. Even if we would accept such a thing, Goodman shows, that it simply doesn’t make any sense to speak of it, because our understanding is always dependent on our cognitive systems and, furthermore, that there are in general no reasons to prefer any system in favor of another.
»Moreover, things do not present themselves to us in any privileged vocabulary or system of categories. We have and use a variety of vocabularies and systems of categories that yield different ways in which things can be faithfully represented or described. Nothing about a domain favors one faithful characterization of its objects over others. « (Goodman 1988: 6)
Though here’s not the place to discuss Goodman’s underlaying nominalism, constructivism, relativism and irrealism (for further reading see Elgin 1997, and Cohnitz 2006), I hold it necessary to discuss Goodman’s related notion of the possibility of »conflicticting truths«, regarding general aspects of ambiguity. If we once have seen and accepted, that it makes no sense to speak from »one single, neutral reality that underlays our true versions« and »that (s)uch a reality is neither sufficient nor necessary to explain matters of epistemology or to distinguish true from false versions« (Cohnitz 2006: 191), we have to follow the notion of the plurality of (world-) versions and in the most consequent sense the possibility of »conflicting truths«. Goodman’s solution of the problematics of the existance of contradictory versions, like »the world moves«, or »the world is at rest«, independent from any system-relativity, is to propose the existance of »different worlds« (Cohnitz 2006: 194 et sqq.). I would propose to replace Goodman’s notion of ›worlds‹ with ›realities‹ (Wirklichkeiten), to make it more easy to follow his radical approach. Because it is admittedly quite contra-intuitive to accept the existance of various worlds (in fact Goodman speaks of a variety of versions, he only said that »there are then many worlds, if any« (Goodman 1988: 51). But I guess it is indeed much more part of our experience to observe the existance of ambiguity (multiplicity of contradictory versions) in our personal, social, political and religous world-versions; in our perceptual, intellectual and emotional realities. We don’t even have to evaluate the existance of ambiguity, but i propose to accept its existance, therefore importance to be investigated from an epistemological point of view, and not rejecting it as a pure and fundamentally worthless defect. As art is an important way to express ambiguity and therefore giving possibilities to deal with it and transfer our experience into ›real life‹, we have one fundamental and radical aspect of the relevance and function of ambiguity in art.
Bibliograpy, catalogues, online ressources, further readings