The Duck-Rabbit

In Mitchell's essay, she discusses the arguments that always come with an image that is considered a metapicture. Though metapictures seem to address a certain global concept, they still leave room for interpretation. The self-reflexivity of a metapicture is dependent and subjective to the 'self' currently observing.

All metapictures seem unable to exist, according to Mitchell, without opposing arguments, views, and/or standpoints. Sometimes there is obvious opposition, such as between Gombrich's reading of Alain's drawing of the Egyptians and the second reading that says, "they are shown (against all expectation) to be just like us," (Mitchell 44). While Mitchell discusses the distinguishable levels when talking about Steinberg's image, "one simply has n-levels of nested representation, each level clearly distinguished as an outside to another inside,"(Mitchell 42) the levels of metapictures I found the most interesting and provocative were the ones in the multistable images; the levels that were indistinguishable.

Even from the beginning of his discussion on multistable images, Mitchell shows that there is an argument of the way these pictures are viewed: by way of the "savage" mind and by way of the "modern" mind, (Mitchell 47). These images blend together and create illusions of an image, they call into question what the human mind sees at first glance. These images causes the mind to question if what it sees is truly what is there. I find these images an extreme means of self-reflexivity because the image can change and differ for each observer. "If self-reference is elicited by the multi-stable image, then, it has as much to do with the self of the observer as with the metapicture itself," (Mitchell 48). Does this mean that multistable images separate one group of people from the other? What does it mean if you can't see one part of the image? What if an observer ONLY sees a duck or ONLY sees a rabbit? If the levels are indistinguishable, the mind must make levels. The mind must go outside of its usual realm and create. To me, metapictures are meant to do this. Self-reflexive images should do this. They should cause the mind, the self, to go beyond the norm.

Tying into Deleuze and Guattari, Mitchell speaks about 'hypericons', "They are not merely epistemological models, but ethical, political, and aesthetic "assemblages" that allow us to observe observers," (Mitchell 49). So now, we have rhizomes, hypertexts, and metapictures all in the category of assemblages. Multistable images seem to be, no doubt, assemblages. They pile on layers of images in order to form one single image from one viewpoint and another single image from another standpoint.

"The advantage of the Duck-Rabbit is twofold: (1) it is a weak or peripheral hypericon; it doesn't serve as a model of the mind, for instance, but as a kind of decoy or bait to attract the mind, to flush it out of hiding..." (Mitchell 50). Metapictures challenge. Multistable images contradict the mind. Hypericons causes opposition to be discussed.


Mitchell, W.J.T. "Metapictures." In Picture Theory: Essays on Verbal and Visual
     Presentation. Chicago: U Chicago P, 1994. 35-64, 82.