What is the nature of mental states and processes? Will my mind survive the disintegration of my physical body? These are some of the important questions which have occupied Paul Churchland, Professor of Philosophy at the University of California, San Diego. His many articles and books such as Matter and Consciousness, and The Engine of Reason, the Seat of the Soul concern topics such as mind, knowledge, science and language. His wife, Patricia Churchland, is also a philosopher, and she writes about the same kinds of issues.
Much of Churchland’s most famous work is concerned with what we call the mind-body problem, solutions to which range between two main positions. On the one hand, there are dualist theories, which claim that the mind exists and is different from the body. Most religions hold this position: the soul lives on after our bodily death. On the other hand, there are materialist theories of mind. Philosophers and scientists who hold the latter position claim that mental states and processes are nothing more than states and processes of the brain. This is Churchland’s position. He is a reductionist or an eliminative materialist. This means that the mental can be reduced to the physical, so that the mind does not exist at all (and hence is “eliminated”).
Churchland’s arguments for this position are based on neuroscientific evidence. Accordingly, he aims to reduce speculations about philosophical issues to a minimum. This method is in opposition to “folk psychology”. Folk psychology is our everyday, non-scientific way of talking about ourselves and our mental lives in terms of beliefs, desires, hopes and so on. In Churchland’s opinion, the focus of folk psychology is wrong: the basic unit of animal and human understanding is not the sententially expressible state such as “believes that P”, “desires that P”, “knows that P”, and so forth. Rather, it is the activation of a large population of neurons in the brain. This is one of the most important claims of Churchland’s theory, which provides the basis for the arguments for eliminative materialism, so it requires some explanation.
Churchland basis his ideas on the theory in cognitive science known as “connectionism”. The connections of neurons in our brains are parallel, which means that sensory inputs are processed simultaneously and not step by step. A normal computer does the latter, so that, if one connection is damaged, the whole system goes down. With parallel connections not every connection is important. The information will be processed even if a tenth of the connections are damaged. This is called “functional persistence” or “fault tolerance”. So, for example, the input of the colour red on our retina will reach our visual cortex, even if a lot of neurons and connections are missing.
When we see an object – for instance, a face – our brains transform the input into a pattern of neuron-activation somewhere in the brain. The neurons in our visual cortex are stimulated in a particular way, so a pattern emerges. The patterns differ of course with each face (or other objects), so that we can distinguish faces on the basis of slight differences. The pattern of a face differs a lot from the pattern of, say, a rock. But the different patterns of different faces look a lot like each other. According to Churchland, there is one pattern that represents the prototypical face, which is the average of a lot of faces you have seen. This prototype has no big nose, nor a small one, but a normal sized nose. The same goes for all other features. Churchland shows by this prototype, that face recognition, and by extension all object recognition, is something for which no consciousness is necessary. On a purely physical basis we can recognise objects. This claim is supported by work with a parallel computer system made by Garrison Cottrell’s group at the University of California, which is designed to recognise faces. This computer has, of course, no consciousness and doesn’t know the meaning of the input, but it is still almost 100% successful in recognising faces. As our brains work in the same way, we don’t have to postulate consciousness to explain how we recognise objects.
Another important point about these connectionist networks is that they can recognise faces which are partly covered. This shows that the computer system or the brain is able to complete the picture of the input; that is, that it is able to recognise part of a pattern and then complete it. What we have here is a primitive form of inductive inference, a very important feature of the human capacity to gain knowledge. Again, Churchland shows that this feature can be explained without an appeal to notions such as consciousness or meaning. If he is correct, it is a very strong argument for reductive or eliminative materialism: we can do without consciousness and meaning and still have the capacity to reason.
Of course not everyone agrees with Churchland’s reductionist position, which eliminates the necessity of personal conscious experience. Thomas Nagel is one of his opponents. Nagel claims that the awareness that a bat has of the world has a distinct, subjective feel, and that this is not captured in physical descriptions of the bat. We don’t know what it is like to be a bat, because we don’t have the experiences a bat has. We can extend this argument: I don’t know what it is like to be you, because you have your own experiences. So experience, and therefore consciousness, cannot be reduced to something physical. Frank Jackson and John Searle give similar non-reductionist arguments against eliminativism. The criticism boils down to the thesis that knowledge of the physical system of a person does not give exhaustive knowledge of the experiences that a person has. The conclusion is that conscious phenomena cannot be given a purely physical explanation.
Churchland’s reply to this attack is that everyone has his or her personal neuronal network, his or her own internal resources for gaining knowledge about his or her own sensory activities and cognitive states. These are personal, causal connections with those states that no one else has. But this does not mean that there is something nonphysical about those states. So there is nothing that suggests that there is something about sensory and cognitive states that transcends understanding by the physical sciences. I can know what it is like to be a bat, but I cannot know it in the way the bat does. Churchland concludes that “there is nothing supraphysical, nothing beyond the bounds of physical science here.”
The argument between reductionists and non-reductionists is still going on. Churchland is clearly a reductionist and wants us to join him. He writes in the introduction of “The Engine of Reason, the Seat of the Soul”: “I hope to make available here a conceptual framework of sufficient richness and integrity that will be able to reconceive at least some of your own mental life in explicitly neurocomputational terms.”