Tuesday, January 22, 2013

In the face of a sunset

I've just started studying psychology, and one of my first assignments, for a module on perception, was to observe a sunset and then write 2,000 words about it. Here is the result (a warning: I went over the word limit, and it's about three times as long as my usual blog posts - but it's pure tango philosophy). All the photos are from a sunset I watched at Long Forest Flora Reserve, near Bacchus Marsh, Victoria, Australia.

“Imagine saying, ‘That sunset is interesting.’” (Sontag 2002: 26)

To sit and watch a sunset with the precise intention of observing how visual perception changes as the light changes is to attempt to find a sunset interesting. Rather than getting caught up in, or attempting to capture, the beauty of the sunset, we try to maintain the distance of a scientist. Is it possible? Can one remain scientific in the face of a sunset?

“In the face of a sunset…” A sunset confronts the watcher not as an object that can be quantified and precisely defined, but more like a person, or a face, that calls up, with its shifting expressions, emotions that are hard to explain. Why should changing light, some streaks of colour and luminosity in the sky, the softening of the landscape into velvety darkness, cause such a welling up of feeling?

A Zen master demands, “Without thinking good or evil, in this very moment, what is your Original Face?”

Neuropsychologists have found that a particular part of the brain, the fusiform gyrus (sometimes called the fusiform face area), is especially important to the task of recognizing faces. When it is damaged, the result can be prosopagnosia: ‘face blindness’ or literally ‘face not-knowing.’

‘Face not-knowing’: an expression worthy of a Zen master. And not a bad way to describe the experience of watching a sunset. We do not know the face of the sunset the way we know the faces of our friends, but there is an intimacy in the experience nevertheless, a recognition that is not-knowing. This not-knowing is not a privation, but an opening onto the world – the world of perception.

Natural perception is not a science, it does not posit the things with which science deals, it does not hold them at arm’s length in order to observe them, but lives with them; it is the ‘opinion’ or the ‘primary faith’ which binds us to a world as to our native land, and the being of what is perceived is the antepredicative being toward which our whole existence is polarized.
(Merleau-Ponty 1962: 321-322)

What is antepredicative being? A commentator on the work of Husserl (the German philosopher who founded the 20th century philosophical movement of phenomenology), describes it in terms of “experience that has a certain original meaning in a pregnant sense; it is external sensitive perception.” (Raggiunti 1981: 256)

Phenomenologists argue that all perception takes place within the hermeneutic horizon of the world, and that the fundamental experience of being-in-the-world precedes any ability to objectify and analyse components of our experience using language. We do not arrive in the world as subjects confronted by a multitude of external objects with which we must subsequently build relations of knowledge. Rather we can only perceive and come to know an object (and assign true or false predicates to it) on the basis of always already being within a world from which knower and object emerge together.

In advanced Western culture, dominated by the scientific paradigm of knowledge, we tend to ignore our immediate experience of the process of perception and assume that we live in a pre-made world of objects. We see a particular hammer, for instance, as an object of a certain mass and weight, certain dimensions, made of wood and metal, which we can, as a logical result of all these qualities, use for hammering. Heidegger argues, against this presumption, that a hammer is most naturally experienced in the mode of the “ready-at-hand,” as something for hammering, an extension of the body engaged in a particular task. Only if something goes wrong, such as hitting a finger rather than the intended nail, will the person using it shift to perceiving the hammer as an object, separate from the subject, alien to him or her: “present-at-hand.” In the more fundamental mode, the hammer is not separate from the person who hammers; its identity is inseparable from the activity in which this person is engaged, and a world in which there is some call for hammering.

[The] less we stare at the hammer-Thing, and the more we seize hold of it and use it, the more primordial does our relationship to it become, and the more unveiledly is it encountered as that which it is—as equipment … If we look at Things just 'theoretically', we can get along without understanding readiness-to-hand. But when we deal with them by using them and manipulating them, this activity is not a blind one; it has its own kind of sight… (Heidegger 1962: 65)

Although motor activity that binds us with the world in meaningful activities does indeed have its “own kind of sight,” something we will come to later, vision is the sense most closely associated with the prejudice that maintains that reality consists of objects or bodies, and we stand before them passively as subjects or minds, somehow receiving impressions, sensations, sense data that allow us (mysteriously) to know these objects. 

Taking the theoretical position that we are less subjects than merely very complex objects among the other objects (brains rather than minds) doesn’t make the ability to know any less mysterious, since objects are almost as foreign to one another as they are to subjects. They can hurtle towards one another through space and collide in complicated ways, but how that creates meaning or “external sensitive perception,” or the sense of a world, remains extremely difficult to explain. Phenomenologists would say that it doesn’t just make it extremely difficult – it makes it impossible. Why? Because it puts the horse before the cart. It assumes that there is an object before it is perceived as such. It refuses to see that the formation of objects depends on the process of perception.

As Berkeley says, even an unexplored desert has at least one person to observe it, namely myself when I think of it, that is, when I perceive it in purely mental experience. The thing is inseparable from a person perceiving it, and can never be actually in itself because its articulations are those of our very existence, and because it stands at the other end of our gaze or at the terminus of a sensory exploration which invests it with humanity. To this extent, every perception is a communication or a communion, the taking up or completion by us of some extraneous intention or, on the other hand, the complete expression outside ourselves of our perceptual powers and a coition, so to speak, of our body with things. (Merleau-Ponty 1962: 320)

Merleau-Ponty’s description of the process of perception may be more poetic than those found in contemporary neuroscience texts, but the story they tell is substantially the same, at least if we focus on the first side of the philosopher’s definition. The neurological process underlying our experience of vision can be defined quite accurately as “the taking up or completion by us of some extraneous intention,” so long as intention is understood here not as a psychological state involving deliberation, but rather as phenomenologists use the word, as implying simply a sense of direction or focus. For the phenomenologist, all mental states are “intentional” in the sense that they point to something; they are “about” something. When we think, we think of something (however vague, or abstract it may be, there is always some content to our thought). This principle is even easier to grasp in the case of sight: when we see, we always see something (even if it’s just a streak of colour). There is no seeing without the thing seen.

But analysis of the neurological process of visual perception reveals that the “thing seen” is not immediately given as a visible image. Between the contact of light waves with the cones and rods of the retina, and the formation of a visual image in our “mind’s eye,” lies a complex and as yet only partially understood process of “taking up” and “completion” of the external stimuli by neural activity. Our brains do not receive ready-made images of the world; they construct them. The “thing seen” is not “out there” in the external world; it is a product of contact between something “out there”– the stimulus, or in Merleau-Ponty’s language, the “extraneous intention” – and the complex internal processes we bring to the act of visual perception.

Interestingly, both the scientific term “stimulus” and Merleau-Ponty’s evocation of an “extraneous intention” suggest that the content of vision is not primarily “intended” by us. Although we have some control over our eyes, and can choose to look in a particular direction – we can set ourselves up to watch a sunset, for instance - when we see, it is because a part of the world has directed itself toward us. Some light rays have sped in our direction and arrived on the retina, sparking the processes by which we then go on to make sense of this contact. In this sense, it seems as true to say that in the act of vision, a part of the world focuses on our eye, as to say that our eye focuses on a part of the world.

Explanations of visual perception often compare the eye to a camera, suggesting that the major role for our neurons is to make sure we invert the image on our retina so that we see it “the right way up.” Among neuroscientists, however, this common understanding has long been superceded. The image of the eye as a camera has given way to an understanding of not only vision, but the whole human perceptual system as something closer to a statistical inference engine, whose function is to infer the probable causes of sensory input.

On this understanding, there simply is no image at the level of the retina. The processing of signals that are triggered when photons (tiny “packets” of electromagnetic radiation) contact some of the millions of photoreceptors in a retina takes place in multiple locations in the brain, (notably in the lateral geniculate nucleus (LGN) in the thalamus, which passes signals to the primary visual cortex, and in the superior colliculus in the mid-brain which assists in controlling motor responses such as eye movements). This processing involves complex statistical “calculations” which compare the inputs from cones or rods (which one will depend on the amount of light involved) both with each other and with “statistics” that have been retained from previous visual experience and are used to predict the likely sense of the incoming data. Neuroscientists such as Geoffrey Hinton and Karl Friston suggest that the early stages of processing are used to form predictive hypotheses about meaningful content at higher levels, predictions which are then revised in the light of, as well as used to guide, more complex analyses of the data (see e.g., Dayan et al 1995; Friston 2010). It is from this process of analysis that the brain produces a visual image which is consciously experienced. This image is a highly developed interpretation, rather than a simple copy, of what takes place on the retina.


To describe vision as a process of statistical analysis is to create a new metaphor for vision to displace the earlier one of the camera. This new metaphor has the advantage of helping us recognise something we can discover directly from perceptual experience: that the recognition of visual images involves an unconscious form of interpretation. This is why optical illusions like the Müller-Lyer illusion work. It is also in evidence in the fact that colours appear remarkably different depending on which other colours they border. Such cases suggest that the process of perception corresponds more closely to the probabilistic mode of inductive reasoning than to strictly deductive logic.

This idea is also dramatically supported by the phenomenon of binocular rivalry, which refers to the fact that when a person’s eyes are presented with different stimuli, subjective perception alternates between them, rather seeing them superimposed. For example, if a person is presented with an image of a house in their left eye and a face in their right eye, they will alternate between seeing the entire house and the entire face. Shifts between these percepts are not always clear-cut: “Dominance breaks through in small patches of the visual field and gradually spreads before completely or partially suppressing the competing image (Lee et al., 2004; Meenes, 1930; Wheatstone, 1838).” (Hohwy et al. 2008: 694) This observation is nicely explained by the notion that the brain interprets data from the visual organs by processing it according to principles of statistical probability, seeking meaningful interpretations of sensory data by repeatedly making likely hypotheses based on rapid analysis of the available data and revising them under pressure from anomalous stimuli.

The limitation of seeing the brain as a statistical inference engine, however, is that this metaphor still remains within the dominant conceptual framework which assumes that prior to our perception of it, the world is exists in the form of objects which we can come (at least probabilistically) to know. Because it remains within this paradigm, the statistical model of perception cannot begin to explain the mystery of consciousness or the question of how it is that we manage to get beyond the limitations of the subjective to connect with the objective world. It stops short of embracing the possibility that subject and object might emerge simultaneously in the processes of perception, and that these categories are not prior to, but rather depend upon perception as their ground. On this (phenomenological) view, our perception of the object is not merely probabilistic or approximate, it is creative. This is not to say that there is nothing beyond our perception, no world that exceeds and stimulates our experience of it. But it is to say that this “beyond” does not take the form of objects.

There are certain perceptual experiences in which the phenomenological perspective breaks through in small patches, as it were, and “gradually spreads before completely or partially suppressing” the competing scientific paradigm. The experience of watching a sunset might be considered a prime example.

A sunset is, self-evidently, not an object, even if we often try to turn it into one by photographing it, so we can carry it away with us as a beautiful image, captured on our smart phone. Or if we are psychology students, we may take notes in an attempt to analyse the changing experience, and its effect on our visual perception: the intensification and then gradual fading of colours, the introduction of a grainy quality in the colours as if fine black soot were being mixed into them as the rods in our retina become active and the stimulation of the cones diminishes, and finally the emergence of a world in almost black and white, where we seem to feel as much as see movements and contrasts, perhaps evidence that the neurons which feed straight from the optic nerve to the superior colliculus become more dominant as the light dims and rods take over from cones. These neurons are the ones believed to facilitate “blindsight.” This is the ability to register visual input in carrying out motor functions without any conscious awareness of this data. It is commonly demonstrated by people who have lost part of their conscious visual field due to damage to the primary visual cortex. (I would suggest that it is also the “kind of sight” that Heidegger speaks of in relation to the habitual use of tools.) For example, a person may not be able to see an obstacle on one side of their body, but can quite accurately avoid it, just as I was able to walk very competently along a bush track by the light of a crescent moon after the sunset had faded from the sky, even though my conscious ability to see the bumps and potholes of the track seemed somewhat less than adequate to the task.

Such observations and speculations may be genuinely (and not merely fashionably) interesting, and the photographs very desirable, even by standards not dictated by commodity culture, but the experience of watching a sunset is clearly so much more than, and so different in quality from, either the analysis or the artefacts we take away from it. It is antepredicative, before and beyond all the statements we can make about it, although we might say that its beauty “has a certain original meaning in a pregnant sense.” If all perception is a communication or a communion with things that always takes place with the meaningful horizon of a world, this seems particularly evident in the experience of a sunset, as we watch the literal horizon of our world grow darker and simpler as the sky above it comes alive with shifting colours which seem to move simultaneously into our bodies in the form of emotion.

Reflection on what we know and what we still do not understand about the way in which our neurons creatively and vitally contribute to such an experience raises the question of whether the sunset really takes place in the sky or in the elaborate architecture of our neurons. Merleau-Ponty’s answer to this question is perceptive – the sunset I experience exists not in the sky or in my head, but rather in an intimate meeting and intertwining of the two. A sunset, like any perception but perhaps more movingly than most, is “the complete expression outside ourselves of our perceptual powers and a coition, so to speak, of our body with things.”

My Original Face, in a certain slowly unfolding moment, completely expresses itself in a sunset.


Heidegger, M. (1962). Being and time (J. M. E. Robinson, Trans.). New York: Harper & Row.

Dayan, P., Hinton, G. E., & Neal, R. M. (1995). The Helmholtz machine. Neural Computation 7, 889–904.

Friston, K. (2010) The free-energy principle: A unified brain theory? Nature Reviews Neuroscience 11:127-38.

Hohwy, J., Roeppstorff, A., & Friston, K. (2008). Predictive coding explains binocular rivalry. Cognition, 108, 687-701.

Kenji D., Shin I., Alexandre P., & Rajesh P. N. R. (Eds) (2007), Bayesian Brain: Probabilistic Approaches to Neural Coding, The MIT Press.

Knill D., & Pouget A. (2004). The Bayesian brain: the role of uncertainty in neural coding and computation, TRENDS in Neurosciences 27(12).

Lee, S., & Blake, R. (2004). A fresh look at interocular grouping during
binocular rivalry. Vision Research 44, 983.

Meenes, M. (1930). A phenomenological description of retinal rivalry.
American Journal of Psychology, 42, 260–269.

Merleau-Ponty, M. (1962). Phenomenology of Perception (C. Smith, Trans.). London: Routledge & Kegan Paul.

Raggiunti, R. (1981). The language problem in Husserl's phenomenology. Analecta Husserliana, 11, 225-278.

Sacks, O. W. (1986). The man who mistook his wife for a hat. London: Pan Books: Picador.

Sontag, S. (2002). An argument about beauty. Dædalus, Fall, 21-26.

Strand, C. (2010) Green Koans Case 12: The Original Face. tricycle. Retrieved from http://www.tricycle.com/web-exclusive/green-koans-case-12-original-face. Accessed 20 January 2013.

Westheimer, G. (2008). Was Helmholtz a Bayesian? Perception 39, 642-50.

Wheatstone, C. (1838). Contributions to the physiology of vision. Part I.
On some remarkable, and hitherto unobserved, phenomena of
binocular vision. Philosophical Transactions of the Royal Society
(London) B 128, 371–394.


Peter said...

Well, I'm not sure I completely understood all that (my fault not yours !) but I like your conclusion that "the sunset I experience exists not in the sky or in my head, but rather in an intimate meeting and intertwining of the two."

In a very different context, religious studies, much of what you say about the sunset could equally be said about God. A number of times reading your essay I substituted "God" for "sunset" (even though God is not usually "seen" in quite the same way a sunset is) and it still seemed to work ...

So for those us us who believe in God, divine reality exists not out there or in my head but in an intimate meeting and intertwining of the two ... Yes, that makes sense, and supports my attempts to help students to move away from the idea of a "view from nowhere" ... to a realisation that all truth claims are contextual.

Something for me to read again and ponder further - thank you Justine !

Jason said...

HD :-)

Juzzeau said...

Thanks Peter, that's a beautiful image of God. I think you are completely right that these ideas are a way of talking about the spiritual. Or you could say that spiritual language is a more usual way of talking about these ideas.

I think part of why this piece is not easy to understand is because it attempts to discuss (spiritual) experience using scientific and philosophical language that is not often used in this way - so the language pulls away from the ideas, a bit. But no doubt I could also be clearer - because I wrote this for academic purposes, I've left some jargon and lots of quotes in that I would normally take out for a blog piece. I might try to write something more accessible in time.

But also, Merleau-Ponty's approach does take a while to get your head around because it is so different to the dominant understanding. It's well worth spending some time reading his work. I'm planning to read his late work, The Visible and the Invisible, soon - I think this would interest you, as well.

Juzzeau said...

Jason, glad you like it, but where are the comments, professor??

Jason said...

I have nothing to teach you, Grasshopper.

Jason said...

On secound thoughts I do have comments.

1. As you know, I study statistical inference. So you made me feel important! Yay!

2. I was discussing this piece with a friend who's not a philosopher, and I found myself explaining the use of technical terminology like this:

Some words in philosophy are used because they make the user sound clever.

Some are used because they refer to important ideas that other people have written about.

And some are used because they're so much fun to say.

I just noticed that [Juzzeau] uses the word "antepredicative". Try saying that out loud! So much fun!