external world is associated especially with our faculty of vision. The distance this faculty affords us from whatever happens to be under observation is so great relative to how our other senses work that it has encouraged us to conclude (mistakenly) that the break between the observer and the observed is clean. In sharp contrast, Merleau-Ponty (1968: 134) argues that we all see, like Lear's blinded Lord Gloucester, ‘feelingly’: “Between the massive sentiment I have of the sack in which I am enclosed, and the control from without that my hand exercises over my hand, there is as much difference as between the movements of my eyes and the changes they produce in the visible. And as, conversely, every experience of the visible has always been given to me within the context of the movements of the look, the visible spectacle belongs to the touch neither more nor less than do the ‘tactile qualities.’” It must follow that visual perception too works as a crossing function (ibid.): “[S]ince vision is a palpation with the look, it must also be inscribed in the order of being that it discloses to us; he who looks must not himself be foreign to the world that he looks at.”
If the seer is thus continuous with the seen, in looking at its other it is also seeing itself. In the visual domain, Merleau-Ponty describes perception along the lines of mirror-imaging, making self and other reverse—and therefore less than absolute—projections of each other. In his phenomenological account of the development of the child's perception of others, Merleau-Ponty (1964b: chap. 4) makes extensive use of the example of how children respond to their own image in the mirror. He finds (on the basis of psychological research) that whereas at first the child tends to see the visual image of its body in the mirror as enjoying a quasi-existence, it gradually learns to displace the image and grasp the mirror's crucial developmental lesson: “[H]e can…be seen by an external witness at the very place at which he feels himself to be” (ibid.: 129). In the following passage, Merleau-Ponty (ibid.) attempts to lend support to the claim of an original and originary consciousness in which differentiation is less than complete:
Many pathological facts bear witness to this kind of external perception of the self…First, it is found in many dreams in which the subject figures as a quasi-visible character. There would also be phenomena of this kind in dying people, in certain hypnotic states, and in drowning people. What reappears in these pathological cases is comparable to the child's original consciousness of his own visible body in the mirror. “Primitive” people are capable of believing that the same person is in several places at the same time. The child knows well that he is there where his introceptive body is, and yet in the depth of the mirror he sees the same being present, in a bizarre way, in a visible appearance.
Granted, his comparison here of “primitive” people with children and pathological cases is anthropologically exceedingly crude and badly dated. But it makes a powerful difference to the anthropological import of the comparison that he aims to disclose by it, rather than (à la Frazer et al.) a mistaken perception of the world, a recognition that opens on the truth or a prioricity of nondualism, and therewith one that betrays the radical reduction carried out by intellectualization. Merleau-Ponty (1964b: 132) relates that Wallon, the psychologist on whose study he draws here, holds that once the child has learned to reduce the mirror image to an ideal space (that is, to intellectualize it), the image has become what it should be in an adult mind: “a simple reflection.” But, says Merleau-Ponty (ibid.), “there are two ways in which we can consider the image—one, a reflective, analytic way according to which the image is nothing but an appearance in a visible world and has nothing to do with me; the other, a global and indirect one, of the kind which we use in immediate life when we do not reflect and which gives us the image as something which solicits our belief.” In other words, he is suggesting that in fact the child's relatively undifferentiated perception has something fundamentally right about it (ibid.): “[T]he image in the mirror, even for the adult, when considered in direct unreflective experience, is not simply a physical phenomenon: it is mysteriously inhabited by me; it is something of myself.” As Wittgenstein would say, it is not a mistake.
If Merleau-Ponty is correct, then the lesson to be learned when the child comes to better differentiate his mirror image from himself, such that he learns to take that image as an external perspective on himself and thence to see himself in terms of how he may appear to the other, is that, far from being perfectly separate and distinct from his other, in some concrete (but, crucially, always imperfect) way he is his other. Merleau-Ponty (1964b: 135ff.) goes on to cite Lacan's famous psychoanalytic study of the role of the “mirror stage” in the development of the self, to the effect that the child's eventual assumption of the viewpoint taken on by him, as this viewpoint is given in the mirror image, is what makes the self possible. For it is only by assuming an ‘outside’ perspective, that is, the perspective of the other, that a self can appear at all to what is otherwise a mere “lived me” (ibid.: 136). “To use Dr. Lacan's terms,” writes Merleau-Ponty (ibid.), “I am ‘captured, caught up’ by my spatial image…The specular image is the ‘symbolic matrix…where the I springs up in primordial form before objectifying itself in the dialectic of identification with the other.’” Put differently, the immediate me is drawn away from itself and subjected. But Merleau-Ponty (ibid.: 138; 140) is keen to stress that this development is made possible only because the lived me, the pre-eminently (but still not absolutely) undifferentiated childhood me, “is never radically liquidated”: “[W]e must consider the relation with others not only as one of the contents of our experience but as an actual structure in its own right. We can admit that what we call ‘intelligence’ is only another name designating an original type of relation with others (the relation of ‘reciprocity’) and that, from the start to the finish of the development, the living relation with others is the support, the vehicle, or the stimulus for what we abstractly call the ‘intelligence.’” For Merleau-Ponty, then, the intertwining constitutes a bodily a priori, one that, in view of its sensible or experiential nature, is no less synthetic than given.
Reversibility or the intertwining appears to make the self-other relationship sym-metrical, but this appearance of symmetry, although provocatively instructive, is misleading. It is instructive because it points to novel realizations, perhaps the most jarring (and, when one considers the considerable ubiquity of ‘the evil eye’, ethnographically gravid) of which is that the seen as well as the seer must have the capacity of sight. Of course, when the seen is another person, there is nothing outlandish about this condition. But as soon as ‘other’ is used to include all that is not self, then it must be the case that visible ‘things’ as well as persons can return the seer’s look. Merleau-Ponty takes this claim quite seriously, as in the following example of painters (1964b: 167): “Inevitably the roles between [the painter] and the visible are reversed. That is why so many painters have said that things look at them. As André Marchand says, after Klee: ‘In a forest, I have felt many times over that it was not I who looked at the forest. Some days I felt that the trees were looking at me, were speaking to me…I was there, listening…I think that the painter must be penetrated by the universe and not want to penetrate it.’” I take it that Merleau-Ponty intends, against all logic, that the trees are actually looking at the painter as he is looking at them (pace Dillon 1988: 169). Under our received ontology (on which logic proper rests), this claim must be regarded as perfectly outlandish: trees as such do not possess organs of vision. But under Merleau-Ponty's (1964b: 163) revised ontology, in which the animate body is without organs, that is, does not amount to “the assemblage or juxtaposition of its parts,” except in the abstract, the body is at once separate from and participant of the seen. In which case, reasons Merleau-Ponty (ibid.: 164): “It is more accurate to say that I see according to [the seen], or with it, than that I see it. ”
Once one takes up the ontological perspective of nondualism, then even if the trees are not seeing in the strict sense, it is sensible to say in earnest that the trees regard the painter. The strict sense is highly abstract, a very useful but readily misleading meaning constructed by setting aside the lived world, the world in which I and the trees participate in each other, such that the trees can see as a function of me and “[t]hings have an internal equivalent in me; they arouse in me a carnal formula of their presence” (Merleau-Ponty 1964b: 164). In effect, the trees serve as a kind of mirror, and the mirror appears because
