§77. Image and Perception
Material nature is the aggregate of images, to which Bergson adds that perceptions are those same images referred to the action of the perceiving body. This again recalls Epicurus, who identified vision’s external object with the content of visual perception.
Organisms are images, like any other body, with one exception. Whereas inanimate images act and react on all their facets and in all their parts, organisms are images that receive action only in special parts and execute motion only in others. In a commentary on Bergson, Deleuze describes organisms as quartered (ecartelees) images. An organism’s receptive facets isolate certain images from all those that act together in the universe. Influences indifferent to the organism pass through as imperceptibly as neutrinos; others are segregated, analyzed, perceived, and remembered. The living image analyzes movements received and selects movements executed. A center appears amid change when a formerly a-centered world curves in, forms a horizon, and creates the point of view of action. “By incurving, perceived things tender their unstable facet towards me, at the same time as my delayed reaction, which has become action, learns to use them.”8Bergson's thesis on perception is somewhat complex, and I approach it in two phases. First (1), the perceived image is the material, physical image, the tableau of bodies and surfaces locally present; but, second (2), this physical tableau is relative to the perceiving body's potential action.
(1) I look to my left out a window. It is late afternoon and autumn. Across a field at some distance, I see the edge of a forest where a beam of fading sunlight falls on one tree still in foliage. I perceive an image, many images, a tableau, but where is it? In my brain or on my retina? No, the image is extra- somatic, beyond the window, at the forest edge.
The image is in nature, in the material world, the physical universe, right where I see it. The image I see and the tableau at a distance are numerically one and the same entitative image.Hume thought you could refute this just by squinting, which should show you that perceptions are not in the space they seem to be. Bergson has not forgotten the dependence of perception on sense organs and the brain. Sever the optic nerves and we destroy vision. But let us not forget something else. What is a brain? What are eyes and nerves? Images. We cannot say images are products of the brain, because the brain is itself an image, and images cannot generate images. All they do is change in response to other images. Neurology is not a condition on the existence of the images we perceive, being instead part of their field, part of matter, which, as Bergson said, is the aggregate of images.
Materialist philosophers have always tolerated one inexplicable transition. In the subtlest vibrations of the finest particles, still completely physical, they manage to find a first image, the birth of perception. Subtler and subtler, fine to superfine, then finally a quantity of motion becomes quality and image. How that happens, how motions become qualities, has never been explained. James did not expect a breakthrough. “ ‘A motion became a feeling!'—no phrase that our lips can frame is so devoid of apprehensible meaning.” For Bergson too the question is badly formed. Moving bodies, however fine, however subtle, are moving images, and cannot generate new images. Images can transmit movement to our body, and the body-image reacts on images; it is not the source of images but an exchange point, a local center of action.9
Bergson describes images as pictorial. “The object exists in itself, and, on the other hand, the object is, in itself, pictorial, as we perceive it: image it is, but a self-existing image.” I suggested thinking of photographs, but images do not copy anything.
There is no model, no Idea of which they are an appearance. To call images pictorial is to say that they have the qualities they are perceived with, as in a picture of a man wearing red pants the picture itself is red, the red really on its surface. Bergson's images are like that, the colors we see really being on the surfaces we see, belonging to the images as red belongs to pictures of men in red pants.10(2) A few surgical cuts and what is the result? We cannot move a muscle, and we lose all of our senses. For these two effects—sense and volition— to vary together suggests a connection between perception and action. Bergson's idea is that the neurological process of perception formats images according to an evolved, species-specific capacity to respond. In becoming perceived, the image is simplified, stripped of qualities the perceiving organism cannot respond to, cut to fit its virtual action. The difference between an image and its perception is that images are continuous, not fleeting and discontinuous, as perceptions are. Every image is embedded in an environmental economy of images and constrained in how it can change. Perceived images are not constrained in this way. We can look aside, close our eyes, or run through a series of imaginary modifications without changing the image.
The theory of primary and secondary qualities held that perception adds secondary qualities to things that in themselves have only primary qualities. Secondary qualities are projected from the subject to the object, which we confuse with our response to it. For Bergson, perception does no such thing. Perception does not add to the image, it reduces and simplifies it, ignoring anything the perceiver (qua adapted species) cannot act on. Perception thus imposes discontinuity, cutting images to our capacity, reducing the endlessly multifaceted to a manifold on our terms.
The discontinuity of images (bodies) is undermined by any close look. Edges and surfaces are artifacts of scale without substantial reality.
We create a working discontinuity by pre-cognitively isolating the image, simplifying it, suppressing whatever we cannot respond to. This was Gibson's argument that visual perception ignores lots of change in the environment, attending only to discernable invariants that afford action (§68). The forms we see, the shapes, the figures, the lines of discontinuity, indicate the stability and change we expect from images and for which perception prepares us.It merely prepares, being not necessarily what we do but what we could or can do and tend to do, which Bergson calls virtual action. Perception “exactly measures our virtual action (action virtuelle) on things”; the “virtual action of things upon our body and of our body upon things is our perception itself” Natural selection tunes perception to an organism’s potential for action, so that by perception an organism becomes aware of change and prepares an adaptive response. Perceptions are tendencies to act, virtual action, not copies of images but preparations for interaction with them.11
The idea of perception as virtual action is one of Bergson’s best, though also consistent with physiological thought in Europe at his time, and other Darwin-influenced physiologists and psychologists were thinking along the same lines, including William James. Writing in 1881, he says, “The current of life which runs in at our eyes or ears is meant to run out at our hands, feet, or lips,” an idea he describes as “one of the fundamental conclusions to which the entire drift of modern physiological investigation sweeps us.” The structural unit of the nervous system is a triad, none of whose elements has independent existence. The sensory impression exists only to stimulate central processes, which in turn exist only to call forth action, a loop both of whose ends have their point of application in the environment.12
In a work called Origins of Objectivity philosopher Tyler Burge writes, “Perception is not perception if it cannot be accurate or mistaken.” He thinks perception makes a truth claim and is right or wrong, having what he calls veridicality conditions, an idea of perceptual content that I argued against in §69.
The world does not come prepackaged in bodies that we perceive more or less veridically. On Bergson’s alternative, perception is defined by the adaptive discontinuity it introduces. Burge assumes that nature is already individuated, it being up to perception to represent that veridically. For Bergson, introducing discontinuity, not veridicality, cutting out, not representing, is perception’s original contribution. “Whatever be the nature of matter, it may be said that life will at once establish in it a primary discontinuity, expressing the duality of the need and of that which must serve to satisfy it.”13It is thought to be a problem in philosophy to explain the relation of perception to reality. We have these impressions, Hume said, but how do we know they resemble their causes? This problem arises only when we think of perceptions as images “in” consciousness, where we contemplate them and draw fallacious inferences about external causes. If we think of perception as Bergson does, then the images we perceive are located in the environment where we interact with them. Instead of two things, inner percepts and outer bodies, the outer body is an image, which I begin to perceive as my nervous system analyzes it in terms of potential change I can respond to. Conscious perceptual awareness of environing bodies is feeling your own body prepare a response. Only preparing, tending; not representation but virtual action. Perception does not construct the reality of things, as idealists say, or more or less accurately represent them, as materialists propose. The images we perceive are external bodies before and after we perceive them, so there is no problem about how perception relates to material objects. The perceived image is the material object and the relation is virtual interaction.
Another conundrum concerns the genesis of perception, which is a problem especially for materialism. How in a world of mere matter do perceptions start up? Bergson's theory of matter skirts the problem nicely.
Matter, physical nature, is the totality of images, most of which are not conscious, though some, being perceived, fall into interaction with organisms and receive their response. All the movement, all the process, from before we perceive, to the perceiving, and after, is the movement of images. Perception is continuous with the material world we perceive, being just one more interaction, one more exchange of energies.Bergson may seem to neglect the dreams and illusions that preoccupied Descartes and Hume, when we perceive images but there is no outer body. Yet it is a mistake to assume that “if a certain arrangement [e.g., dream or hallucination] produces, at a given moment, the illusion of a certain perception, it must always have been able to produce the perception itself.” To assume that is to forget about memory. The images of a hallucination or dream could not arise merely by imagination; they presuppose the memory of veridical perception and do not disclose perception in its original state. If we had never enjoyed normal vision we would not hallucinate. If we had no memories we would not dream.14