However, if this is true, my reason for believing Dawkins can only be that I am genetically programmed to do so, it is predictable, I obey causal laws. So I do not believe him because he is right and has persuaded me, but because it was always going to happen, granted my biochemistry. Some argue that this circular argument invalidates a fundamental assumption of Dawkins’ and shows it to be incoherent: it does not make sense to adopt this position.
The views of the Materialists such as Richard Dawkins leave no room for an immaterial, separate soul. His commitment to the empirical method is presumed to provide the only true source of knowledge. ‘Soul one’, the Cartesian ghost in the machine, is an impossibility. I shall set out how he justifies this below, but is this a coherent view? This post will argue not. Firstly, materialism says that all that exists conforms to physical laws, including causation, and so, using those laws, we can predict the future in an increasingly determinist way as we understand better the science. Liberty in our thought-life, and even of a separate self, is an illusion because in reality thought, feelings, ambitions and desires are all electro-chemical phenomena, increasingly observable through neuro-imaging. Thought is a complex neuronal activity of stimulated synapses responding to sense-data. “I” am an abstraction from this process, as we shall see below. Now, for Dawkins it must be said that this makes our sense of an independent self and our ‘qualia’ (properties and experiences experienced by a person) into materially explicable brain responses, and so they can be examined and ultimately predicted: our brain is simply a data-crunching machine and our freedom is a useful illusion. Ultimately all behaviours will be predictable because they are determined by our brains and hence by our genes (and memes – see below). However, if this is true, my reason for believing Dawkins can only be that I am genetically programmed to do so, it is predictable, I obey causal laws. So I do not believe him because he is right and has persuaded me, but because it was always going to happen, granted my biochemistry. Some argue that this circular argument invalidates a fundamental assumption of Dawkins’ and shows it to be incoherent: it does not make sense to adopt this position. Secondly, many look at the computer model of consciousness he embraces and argue that it is flawed in its own terms. His view is known as Functionalist, namely, he believes that mental life is a product of the physical life of the brain and has no meaningful separation. It is ‘uncannily computer-like’ (River Out of Eden) because bytes of programmed self-replicating data have evolved to co-operate for the ‘survival machines’ that we are, even to the extent of creating the mind mirage as a useful fiction for planning our own survival. Our brain is hardware, our mind is software, a product of programmes set deep in our genes and playing out a virtual life of their own. Our brain is data-crunching binary-coded neurons that have learnt to survive together and in doing so now project a mind in order to handle their own functioning better. The mind and its sense of self or “I” has no real existence and no real influence, no more than an image on a screen – neuro-science has long since shown that the effects of a putative soul can be replicated by electrical charges, drugs and other biochemical research. Many argue that this view does not add up, not just because it is contradicted by other evidence, but in its own terms it is incoherent because it ignores the very content of those thoughts and mind: qualia, perceived experiences, the individual sense of developing personhood and the indeterminacy always of being able to ‘change one’s mind’. In other words, what it is like to be me. The genome cannot give the content of our memory. In ignoring this Dawkins has shown a prior commitment to his theory, often eloquently expressed and using Biblical imagery (e.g. he starts River out of Eden by referring to Abraham), and then he attempts to squeeze the evidence into it. This is the very opposite of the empiricism he claims to follow, and so this is an incoherence in his argument Thirdly, Dawkins has borrowed the idea of a ‘meme’ and developed it in order to fill in gaps in his argument. The meme is virus-like transmitter of culturally accrued understandings, fashions, preferences and behaviours which have been found to have survival advantages and so have survived and reproduced. Memes are bytes of cultural data which have worked with the genes in our evolution to produce increasing consciousness and even a sense of a separate self that can aid our survival better. One meme-myth is religion, another the soul. The Cartesian soul no longer has any survival value for us, though, and so should be rejected: neuro-science is now explaining our identity and functioning and providing improvements for our inner life much more than a separate soul could ever do. Soul two is this modern Functional soul, sometime also understood in terms of Identity Theory. The perceived incoherence here lies in the old rhetorical problem: it is one thing to state a view and make a claim, it is something else to provide proper explanation. Proper explanation is not a matter even of listing one’s reasons for holding a viewpoint, it is instead a matter of persuading by well interpreted evidence and successful refutation – not mere mockery – of other views. Scientifically a ‘meme’ cannot and does not exist, and so it contradicts an empirically-based argument for a ‘soul two’ i.e. consciousness explained by neural networks. In stating Functionalism and Identity Theory so loud and clear Dawkins has brought out into the open much important debate and has challenged those who prefer property dualism (Greenfield) or substance dualism (Hick, Swinburne, Ward) to fashion better arguments. But despite this important function in continuing debate over Body and Soul, many suggest that Dawkins’ own theory cannot stand as it is incoherent.
A formal argument in logic that is formed by two statements and a conclusion which must be true if the two statements are true. Last week Richard Dawkins got in trouble on twitter for trying to show how one type of syllogism works (https://richarddawkins.net/2014/07/response-to-a-bizarre-twitter-storm/). I don’t want to get into whether he was right for saying what he did - but it was interesting that a lot of people criticising him missed the whole point of his original tweet - to demonstrate the power of the syllogism! So what is this form of logical reasoning and why is it useful to philosophers? At its most simple a syllogism is two statements that relate to one another and lead to a logical conclusion. A basic example is: The beauty of a good syllogism is that the conclusion must be indisputable if the statements are correct. Most A-level students first come across syllogisms when they study the Ontological argument. Anselm, when he wrote Proslogion he lays out his argument in chapter 2 in a fairly verbose way:
Quite complicated! However, it can neatly be summed up in this syllogism:
Now, having simplified the argument into a syllogism, we can then go on to test and discuss the logic and the reasoning more easily - this is why they are so useful!
|
Archives
June 2016
Categories
All
|