Reason in the Machine
The idea that human beings are at least in part mechanical is not a new one. More than 350 years ago, Descartes observed that some behaviours and bodily processes might follow principles similar to those by which machines operate (4). The lens of an eye, for example, refracts light in a similar way to the lens of a camera and the two can be usefully compared to understand their mutual mechanical properties. What is new, however, is the idea that sensations, thoughts and emotions are also the products of mechanical principles.
Descartes recognised that human beings have a capacity for culture, language and free will that sets us apart from mere machines. In a famous experiment, Descartes isolated the lens of a cow’s eye and looked through it to see an inverted world. He recognised that the image our retina projects would also be inverted but we do not see an inverted world. Thus Descartes argued that our perception is not a mere reflection of physical reality but an interpretation based in reason.
Many puzzles remained regarding exactly how the mechanical function of the body relates to sensation and how perception might be derived from what the senses delivered. Descartes rather notoriously explained sensation and perception through divine intervention: God translates physical activity into sensation, and God also provided divine minds that are able to array and order sensation into coherent perception.
Modern neuroscientists rightly reject the need for a divine mind but, in the process, they also reject Descartes’ distinction between machines and humans. Unfortunately, modern neuroscience has no better solution than Descartes for how the mechanical activity of the brain is translated into sensation and how sensations are cohered into perceptions.
Getting feeling out of the brain
To explain sensation, most neuroscientists suggest some sort of identity between it and activity in the brain, or claim that it emerges out of the complexity of neural activity, or some combination of the two. For example, an experience of ‘blue’ involves short wavelengths of light transduced into chemical and electrical activity in the cells of your eyes, your optic nerve and your brain. For the identity theorists, the chemical and electrical activity in specialised cells, nerves and brain areas is the experience of ‘blue’. Such a position is not easy to defend because blue is described and experienced as colour (with a particular brightness and hue) and not as chemical and electrical activity (with a particular pattern and location in the brain). It is reasonable to point out that chemical and electrical activity in particular cells and parts of the nervous system is often accompanied by the sensation of blue. But that is not the same as arguing that chemical and electrical activity in particular cells and parts of the nervous system is blue.
The idea that the experience of blue is a natural, direct and unavoidable product of that chemical and electrical activity in particular cells and parts of the nervous system is more plausible, but it’s still problematic. It remains highly uncertain as to how the experience of blue, or any other sensation, falls out of the physical activity. For example, John Cacioppo and Jean Decety, two neurobiologists at the University of Chicago, describe one possibility:
‘[T]he mental activities surrounding consciousness, perception, and thought are the result of columns of neurons stacked next to each other, forming elemental circuits that function as a unit, with multiple assemblies of such units within the brain working together as a network of distributed, sometimes recursively connected, interacting brain regions with the different areas making specific and often task-modulated contributions.’ (5)
That’s an awful lot of technical words that don’t really add up to much explanation. The language implies precision - neuronal activity is unified, connected and modulated – but that precision lies within a vague ineffability. How does the unified, connected and modulated activity create ‘blue’? The answer is simply unknown.
Let us, however, for the sake of argument, allow that ‘unified, connected and modulated activity create “blue”’. Somehow, blue, and all other basic sensations (or ‘qualia’), are directly delivered by nervous system activity. That still leaves us a long way from a unified perception. Right now a boundless number of stimuli are hitting your senses – lights, sounds, odours, pressures, tastes – that should generate a cacophony of fleeting experiences as one sensation, or mush of sensations, gives way to the next. But one thing after another, after another, after another, and so on, does not make a unified perception. Unified perception requires a point of view, a perspective, to extract one body of stimuli as separate from another. Physical entities, including the brain, do not have a perspective. As a physical entity, the brain just is, just like all other physical entities.
To get around this problem, neuroscientists tend to use a variety of tricks. One trick, which we have already looked at, is to use vague language that implies a highly technical explanation without delivering any mechanism. For example, a neuroaesthetics paper might state that observing a painting by Caravaggio ‘is associated’ with different neural activity from that which occurs when observing a painting by Turner. The implication is that the different associated neural activity accounts for the different aesthetic reaction to each painting. Yet that very implication will remain unexplored.
In reality, observing any two different things will be associated with different neural activity, simply because the things being observed are different. The different associated neural activity provides no explanation for the different experiences. ‘Associated’ neural activation is not an explanation of anything; it is just a description of some part of what happens in the brain when different things are observed. Terms like ‘associated with’, ‘based in’, ‘related to’, and so on, are used because they are generally true (there are always associations and relations to brain activity) while also implicitly delivering a stronger message of brain activity being the important and causal factor. The reader (and likely the writer) tends to automatically assume that the brain is the most important component, just as you might assume legs to be the most important component when reading the sentence, ‘legs are associated with walking’. Of course, legs are associated with walking, just as brains are associated with perceiving, but legs do not exactly cause you to walk. You walk to get somewhere, and you will use your legs to achieve your goal, but you do not walk because you have legs.
Another trick is to smuggle perspective into the description of brain activity; a trick that professor Raymond Tallis has called the fallacy of the misplaced epithet. Misplaced epithets are common, such as when we innocently ask what time the clock says it is (clocks don’t tell time, we do) or when we less innocently berate a computer for refusing to recognise an important file (computers don’t recognise). Neuroscientists employ a wide range of misplaced epithets, including suggestions that nerve impulses deliver ‘signals’, contain ‘information’ or are ‘motivating’, and so on.
Nerve signals do not, and cannot, provide these things because signals, information and motivation require a point of a view, a person, in order for the signal to be referenced, the information to be understood and the motivation to occur. By using such terms, neuroscientists smuggle perspective and subjectivity into nervous tissue and inadvertently ascribe the properties of experience directly to the brain. The purported aim, which is to explain these things as an emergent property of physical activity, is thereby seemingly achieved. In essence, however, the tendency to ascribe experience directly to nervous tissue is not very different from Descartes’ suggestion that perception falls out of the mechanical activity of brains because God divinely arranged it to be so.
A related trick to the misplaced epithet is misplaced order and direction. Driving home last night, I observed a car with a crumpled front end and inferred an accident involving the same car in a rather different state. It is tempting to argue that the crumpled car contained a record of the non-crumpled car that existed before the accident. But such an argument is nonsense. A crumpled car is just a crumpled car. It is not a record of its previous non-crumpled state. Only to an outside observer (me, in this case), who is highly familiar with non-crumpled cars and the crumpling effect of high-speed collisions, is the crumpled car any indication of a previous non-crumpled car. Physical objects qua physical objects do not have a history, they do not contain memories; they just are.
Thus, when neuroscientists suggest that activity in one region of the brain drives activity in another, they are, once again, smuggling subjectivity into a place it doesn’t belong. It is only from a point of view, from a perspective, that one brain state can be seen to be about another brain state. But perspective is not inherent to different brain states, regardless of how temporally interlocked they might be. The ordered activity of the brain cannot be assumed inherently to deliver the ordered nature of subjectivity, because ‘order’ is itself an interpretation of physical events that relies upon subjectivity.
By now you may be wondering where all this negativity is taking us. It takes us nowhere except to a place where all the neuro-prefixed disciplines and topics can be appropriately ridiculed. They provide no better basis for understanding than the non-neuro-prefixed disciplines and topics. At best, the addition of ‘neuro’ provides novel descriptions and perhaps new ways of considering those disciplines and topics. At worst, the addition of ‘neuro’ mystifies and obscures more appropriate ways of considering those disciplines and topics.
A different approach
More positively, attacking neurotrash might force a consideration of what it really is that makes us the kinds of creatures we are. If the answer to human experience, behaviour, subjectivity, consciousness and so on does not lie in the brain, then where does it lie?
There is no doubt that there is something peculiar about human experience. Conscious experience, as I have discussed, refuses to be reduced to physicality. Indeed, it somehow seems to transcend mere physicality. Yet, while that is peculiar, it is not so peculiar as to warrant a descent into mysticism. Many things transcend mere physicality because they rely upon a relationship between physicalities that transcends the immediate physical properties of each individual element.
Let us take the simple example of cows and grass (borrowed from George Herbert Mead (6)). We know that grass is food for cows because cows eat grass; the foodiness of grass is revealed in its consumption. If nothing ate grass then grass would not be food. But the foodiness of grass is not a physical property of grass and is not a physical property of cows. The foodiness of grass is only revealed in the relationship between cows and grass (namely the consumption of the latter by the former). That relationship is real, but not exactly physical.
The wonder of human experience similarly reveals itself not through direct physical expression, but through relations that are not exactly physical. The newborn infant is not merely displaced into a world of stuff but is, importantly, put into direct contact with conscious agents who use abstract symbols and representations that make their inner world public. Thus there is the scope for triadic relations between infant, caregiver and physical objects that immediately transcend the separate properties of the infant, caregiver and object. Through triadic interactions, the spontaneous behaviour of the infant (to grasp, gawp and gulp, among other activities) is linked not just to objects (stuff) but also to a third-party perspective. Thus the infant is provided the opportunity to view, hear and touch the world through another (‘daddy sees that I grasp for the cup’) that facilitates directed activities such as protodeclarative pointing and gaze monitoring, which are embedded in a shared attention that goes beyond the immediate physicality (7).
The infant, caregiver and object do not merely bump together like billiard balls, but are, instead, revealed and transformed within the relationship. Initially, the infant grasps towards an object; yet the grasp is realised not by the infant, but only by the caregiver. Later, reaching by the child becomes an expression of intent, a meaning, an idea directed towards the caregiver. The act of reaching calls out a meaning known to the caregiver and now significant also to the infant. This entire exchange contains within it the beginnings of a mutual experience and language.
The introduction of language is especially useful in drawing the infant away from the mere spontaneous expression of his existence and into a world of thought, reflection and self-control. Words provide many opportunities for reflection and self-control that are otherwise difficult or impossible. Words allow infants to single out separate elements and overcome the natural structure of the sensory field to form new (externally introduced, culturally specific and negotiable) structural centres. It also becomes possible temporally to separate thoughts about the world from immediate sensory experience. In this way, infants can break free from the cacophony of stimuli and from the immediacy of sensation that would otherwise drown consciousness in a totality of the here and now (8).
The infant certainly requires a functioning brain in order to take advantage of triadic relationships, abstract representation and language, but the brain is not sufficient, in itself, to generate conscious experience. That is because the language of consciousness, meaning and representation are not the products of individual brains. Rather, they are the products of a set of organised relations existing or subsisting mentally in a field that is social rather than neural. Attempts to find conscious experience in individual brains or neural circuits inevitably means avoiding the social relations that are necessary for conscious experience. Consequently, conscious experience is either negated (there is no such thing as consciousness or free will) or viewed as mystical (God does it).
Escaping neurodeterminism is vital; it means escaping the idea that human beings are just another machine. Machines, and all physical objects, cannot provide the necessary point of view to be conscious. We need to develop an understanding of consciousness as involving a nexus of relationships between people and being an objective product of that nexus. Consciousness is real, in just the same way that grass really is food for cows, but consciousness cannot be found in the brain. Consciousness is not something substantive and cannot, therefore, be physically localised.
Stuart Derbyshire is an associate professor in the Department of Psychology, National University of Singapore (NUS) and A*STAR-NUS Clinical Imaging Research Centre.
For permission to republish spiked articles, please contact Viv Regan.