There are two interrelated issues that need to be examined in connection with the self. Firstly, whether the self exists and secondly if it does, whether it is in the brain.
All the materialists (e.g. Crick, Dennett and others) argue correctly that there is no audience, no homunculus (an ‘observer') within the brain. There has been a consensus that no single cell or group of cells is likely to be the site of conscious experience. So, in order to remain faithful to their ideology, they must reject that the one who is aware, who is experiencing, exists at all, despite the fact that it contradicts common sense. Note that there is not either empirical or rational support for such an assertion - it is purely ideological. In fact, there is much support that the self that is experiencing does exist (although not necessarily within the brain or body).
The sense of self is universal to all human beings and possibly other life forms. A unique (subjective, first person) perspective that we all have and machines do not, however ‘clever' they are, indicates its existence. The self is sometimes dismissed as an illusion and compared to perceptual illusions. However, as already pointed out, there is a fundamental difference between these two. The sense of self is not based on perception, and therefore cannot be an illusion. It is a phenomenological experience not mediated by the senses that can trick us. Dismissing such experience as an illusion is on the same level as dismissing that reality exists (‘maybe we are all dreaming...'). Hume declared that the self does not exist because he could not find it anywhere. However, this conclusion does not necessarily follow from the premise. Self cannot be directly aware of itself because it is the source of awareness, like a torch that can illuminate everything except itself, or an eye pupil that can enable seeing many things, but not itself. The self cannot be ‘found', but it can be recognised as a source - something that is not in the mind, and yet is necessary in order to have the first person perspective. It can be compared with the conductor of an orchestra (although, admittedly ‘the orchestra' in this case is often ruled by other factors rather than the self). When listening to an orchestra the conductor cannot be heard, yet shis role is indispensable. Hume is mistaken to regard the self as ‘nothing but a bundle of different perceptions'. Perceptions - that is, thoughts, sensations and so on, is what the self is aware of (they are the materials of awareness), so it cannot be reduced to them. Popper, writes:
One might be tempted, under the indirect influence of Hume, to think of the self as the sum total of its experiences... But it seems to me that this theory is directly refuted by the memory experiences... At the actual moment at which the memory delivers something to us, neither the delivering memory nor the object that it delivers to us is part of our selves; rather, they are outside of our selves, and we look at them as spectators (thought we may be active immediately before and after the delivery) and, as it were, watch the delivery with astonishment. We can therefore separate our conscious experiences as such from our selves. (Popper and Eccles, 1977, p.488)
Some scholars (e.g. Susan Greenfield) consider the self an emergent property of the whole brain, but this does not seem to work either. If the brain and the mind are the same, and the self is distributed in the brain, it should be distributed in the mind too. However, this does not correspond to the common experience. Take, for example, an ordinary dream. In a dream there are usually many characters that, from this position, should be all the result of neural activity. Yet the self identifies with only one character (you know in a dream who is yourself and who are others). If the self is the sum of all the processes, it should equally identify with all the characters which is clearly not the case - a single point of view is always present. Moreover, as already discussed above (p.108-109) if the self is identified with a whole (rather than being a distinct element) it would be reduced to a passive ‘observer', an epiphenomenon.
Empirical evidence also supports the idea of the self. For example, split brain surgery (the removal of the corpus callosum that links the brain hemispheres) conducted by Sperry in the 1970s and followed up by Victor Mark and others, show that even when the two sides of the brain are separated, the person is not usually aware of it and acts normally. There is a sense that the two hemispheres still form an integrated entity, although the information they share is minimal. Brain stem and cross-perception (perceptual stimulation of both hemispheres at the same time) may play a role, but they are not sufficient. It is observed that split-brain patients can perform complex activities such as playing the piano that require a high level of synchronisation between left and right hands (and therefore two hemispheres). This integration must happen somewhere else. Only sometimes or under experimental circumstances does it become transparent that the hemispheres do not communicate directly and may even conflict each other. It would be premature, however, to draw a conclusion from such instances that split brain leads to split self, especially if it is taken into account that these experiments rely on short term memory and verbal reports that heavily depend on the brain (in the latter case, mostly on one hemisphere). Even ordinary people often have internal conflicts (one part of the person wants to go out while the other wants to stay in), but this is not to say that a single perspective, from which a whole conflict can be experienced is not retained. Quite the opposite, that the self can be aware of the processes in both hemispheres (although not necessarily able to formulate them), identify with one, or shift between them, indicates its relative independence from the brain.
Blind sight experiments point in the same direction. It is observed (in animal and human subjects) that if a segment of the occipital cortex is damaged or surgically removed, the subject is not aware of a part of shis visual field (although shis visual apparatus work properly). However, if asked to guess what is in that part of visual field, they guess correctly in a number of cases far above the statistical average. This means that neuronal processes cannot be identified with awareness. It is more likely that the occipital lobe plays the role of a relay station, producing waves accessible to awareness. Scott, a scientist with a special interest in consciousness, concludes:
...the split-brain and blind sight experiments, among others, provide objective evidence for the existence of a mental monitor that might or might not be in operation during a particular act of visual response. (1995, p. 162)
After all, if it is accepted that we can be aware and act intentionally, it must be somebody or something that is aware and intends (and if these are the properties of a non-material aspect, the self too is not likely to be material).
The so-called binding problem is also relevant in this case. Everyday experience and experiments on seeing and hearing demonstrate that there is a unifying quality to consciousness. In other words, perceptions, beliefs and attitudes ‘hang together'. All our experiences are closely related and integrated not only with past experiences but also with our actions, expectations, theories, evaluations. However, the brain contains no corresponding kind of unity. Binding or the unitary character of the experience, is not reflected in its structure or functions. Constitutive parts that make up the form of an experience (a movement, colour, shape, sound, smell, texture etc.) are processed in different parts of the brain. Considering this segregated nature of the brain and the relative absence of multi-modal association areas in the cortex, the question is how neuronal inputs are synchronised and overlaid to form a single unified and meaningful perception. A simple example: the image on the retina is processed in over twenty different areas of the cortex, each of them dealing separately with specific features of the image. Neural machinery that could recombine the output of these specialised visual feature detectors is not found in the brain. So, it is unlikely that these ‘point-events' are fully integrated again by purely physiological activity. Eccles writes:
...all the time we are learning more and more about feature extraction neurones and how they come to make more and more complex patterns but never does it get beyond the stage of showing us more than little flashes of simple geometrical fragments to which each cell is responding specifically. (Popper and Eccles, 1977, p.533-534)
Of course, there are several levels of integration: the integration of a simple image such as the shape of an object; the integration of the shape with colour and movement; the integration of a complex image (such as scenery); integration with abstract elements (e.g. a name). Indeed some binding that is dependent on intrinsic properties of the stimuli (such as direction of movement, time, edges, intensity, etc.) may occur even on a sub-neuronal level, on the level of neurons, and some binding probably also occurs on the higher level of neural organisation (cortical regions). But this is not enough:
All we were working with there are patterns of impulses signalling progressively more complex features. There has to be an interpretational read-out. This is what we believe to give us a unified picture and it is a picture involving all kinds of features such as light and colour and depth and form. ( ibid., p.534)
This integration is, therefore, not an ‘objective' (as, for example, the integrative character of a computer programme), but ‘subjective' one that can only derive from the focal point that keeps it all together - the self. In order to preserve materialistic dogma and avoid this simple common-sensical explanation, a number of complicated theories are employed to explain the binding problem (Chaos theory, Quantum theory, Object Template Constrained Feature Processing theory, etc.). Yet, with all that armoury none of them can account for all the cases of everyday visual perception, let alone other types of binding commonly experienced. Most of these theories boil down to variations of Hume's claim that the elements of perception are bound by spatiotemporal association. However, Kant already pointed out the shortcomings of associationism. We must actively bind the various features of objects together, so that what we see are the synthesised constructions of our world. Therefore, intentionality (to use Brentano's term) must be involved. The relatively recent empirical research, such as the work of Ann Treisman (1986, 114-125) on illusory conjunctions (‘mistakes' in binding), also indicates that attention, which can be defined as intentional awareness, is essential to binding.
In any case, on the level of brain processing, waves or oscillations that are a product of neural activity are likely to play a role. Indeed, research suggests that assemblies relatively far from one another synchronise their activities at a frequency between 40 and 80 Hz (the gamma range). They become phase-locked oscillations. However, this does not solve the binding problem as some scientists are tempted to declare. Nunn (a neuroscientist himself) writes:
...all that has been achieved is to show that temporally determined groupings are important to brain activity as well as spatially related groupings... This is an important step forward but it does not obviously get us any nearer towards accounting for awareness. (1996, p. 35)
Commenting on Libet's experiments on temporal discrepancy (that we live up to half a second behind the times) Nunn concludes that ‘the main casualty of accepting the obvious interpretation is any hope that coherent EEG activity on its own can account for the binding problem in relation to awareness since epochs of coherent activity are reaching their end before awareness occurs' (ibid, p.42). Furthermore, the discharges were also found in a wide range of states not connected to stimulus interpretation, so synchronous oscillations may be necessary, but cannot be sufficient for feature binding. For the final level of binding these phase-locked oscillations need to be selected and ‘picked up', for which awareness and intent are essential, and consequently the self - as their source. Only then they can be connected in a meaningful whole:
This read-out by the self-conscious mind involves the integration into a unified experience of the specific activities of many modules, an integration that gives the pictured uniqueness to the experience. (Popper and Eccles, 1977, p.388)
This can also explain why it is possible to use the same group of neuro-connections in presentation of different objects. Many other phenomena point in the same direction. For instance, a changing brain would not be able to perceive a change without a non-changing element and yet, there is not an unchanging part of the brain. Thus, it is proposed that awareness, intent and the self enable particular connections out of endless possibilities (even within the restrictions of time-space associationism). Some neuroscientists, such as Sperry and Doty, who are not dualists, also acknowledge that integration seems to be best accounted for in the mental sphere. Eccles concludes:
...it has been impossible to develop any neuropsychological theory that explains how a diversity of brain events comes to be synthesized so that there is a unified conscious experience of a global or gestalt character. The brain events remain disparate, being essentially the individual actions of countless neurones that are built into complex circuits and so participate in the spatiotemporal patterns of activity... the experienced unity comes, not from a neuropsychological synthesis, but from the proposed integrating character of the self-conscious mind. (ibid., 362)
This sort of binding is not only necessary for perception but also for a meaningful, intentional action. At least some actions must be the result of an interaction with something outside the system producing these actions (i.e. you cannot move a boat by blowing into its sails if you are in the boat). Admittedly, this interaction could be between the brain and the environment, which is why behaviourists adopted the belief that we are completely socially determined. Yet, this is not the case. Every person is (to a degree) an agent. The movements of physical objects may not involve purpose (as Aristotle thought), but purposeful, teleological causation is common to human beings. We are often motivated by the future, an end result. In other words, many actions are meaningful - and such a teleological pull is different from a conditioned push. However, any purposeful activity requires something that can grasp the causal relationship between an activity and its results. Such an element needs to be, as it were, outside any specific process that contributes to the action. An analogy with a factory as a complex system may be appropriate here. The production of the factory may consist of many relatively discrete and specialised processes. Each of them is part of the whole, but none of them determines the whole. Only something that can grasp the whole and its relation to the external world can determine a meaningful direction. This, of course, does not require interfering with or being aware of all the individual processes. Particular units of the body, such as the digestive system, can act independently, but those actions are only re-actions, rather than a directed activity that involves the whole body. The same applies to the brain with its modules that have specific functions - in most cases these modules will carry on unabated by the choices a person makes, and yet they serve them.
All the above suggests that the existence of the self makes more sense than otherwise, and that this universal human experience can be vindicated by empirical findings and reason. However, as already mentioned, everybody agrees that the self cannot be found in the brain. And if it is not in the brain, it must be elsewhere. That the self is located in other parts of the body is even less likely (which is apparent from spinal cord injuries that lead to paralysis but not a loss of self). So, concluding that it is an element of non-material reality seems inevitable. An attempt can be made now to conceptualise that to which the self, awareness and intent belong to.
- . Nagel writes: ‘It is impossible to exclude the phenomenological features of experience from a reduction in the same way that one excludes the phenomenal features of an ordinary substance from a physical or chemical reduction of it - namely, by explaining them as effects on the minds of human observers... The reason is that every subjective phenomenon is essentially connected with a single point of view, and it seems inevitable that an objective, physical theory will abandon that point of view' (1981, p.393).
- . This applies even in the extreme cases of so-called split personality, which are most likely the result of an impermeable segmentation of that with which the self identifies. Sufferers of this disorder still maintain a unique, first-person perspective that enables them to become aware of or dis-identify with some personalities.