A few years back, I wrote a book called Falling for Science, a response to a year spent at The Alan Wilson Centre, a research organising specialising in molecular evolution. The book was a response to my first encounter with the philosophy of science, and was my way of trying to make sense of the way science and story telling are necessarily intertwined. The conclusion, perhaps banal, was that we make sense of the world both by constructing models of our physical experience, and by using story to interpret those models, so freighting them with personal meaning.
Next week I’ll be giving a talk to The Sea of Faith conference, where a theme is the relationship between brain science and spirituality. I want to look at the way certain scientific discoveries constrain our meaning models, and specifically to make the case that, if we are to accept evolution, we are led to the conclusion that the human mind has no reliable access to moral truths.
That’s a reasonably challenging view; for most of us, the idea that some things are just right and wrong, and we know what they are, dammit, is instinctive. And, for most of us in the modern age, evolution stands uncontested. So, if these two cherished views are incompatible, that’s going to cause a problem. Not surprisingly, a great number of people, amongst them eminently qualified philosophers, disagree with me. Here then, in truncated form, is the case against moral knowledge, along with a quick examination of a counter-argument, from The Stanford Encyclopedia of Philosophy, which was provided by a correspondent, Darrell.
Two fundamental scientific models are being assumed. First, that complex design in nature is the result of natural selection. Given that the brain is the most complex piece of design we know of in nature, the argument is that it slowly developed its characteristics over time, in response to the differential survival rates of particular genetically determined variants. Second, that the laws of quantum physics are reliable. That is, that the physical sub-structure of the brain behaves according to probabilistic determinism. A particular state of the brain exists as a probabilistic function of its state in the previous time frame. In other words, we don’t have non-physical entities exerting any systematic influence on physical entities.
The moral argument is a conditional one. It doesn’t say, we must accept such science, but rather says, if we do, we are left concluding that the brain does not reliably access moral truths (it may hit them by random, from time to time, if they do indeed exist). In form, it is essentially a plausibility argument. It says, if we know of no mechanism by which a physical occurrence could occur, and have no evidence that it is occurring, we should conclude it does not occur. Think of this in terms of the claim that at night, while you sleep, your kitchen furniture levitates. A strong scientific argument against, that most would accept, is that we know of no mechanism by which a table or chair could levitate. So it is with this particular argument.
What then, is the equivalent of levitation? Consider a moral judgement. You see a child being beaten and cry out ‘stop, that is wrong!’ We can think of the time between the beating being observed, and the utterance being offered, as a sequence of physical states. First, light of the image hits the retina. Next the light information is converted to electrical impulses and so forth, right through to the brain state required for the utterance to be primed and offered. Physics suggests that every physical state is probabilistically determined by the previous state. In other words, the information at state one is sufficient to explain state two. Nothing uncaused happens within the physical brain, beyond the random element of the quantum equations (which being random, is without cause or purpose).
So, if the final utterance is reliably moral, then in some way, the design of the brain has been set up such that this ‘machine’ is calibrated to true moral values. However, evolution tells us that the design is the result of pragmatic forces. In other words, our particular moral instincts have evolved over time in response to the impact they have on genetic reproduction. Hence a fierce desire to protect one’s own children, that does not extend with the same ferocity to the children of strangers. We have no reason to believe that such pragmatic needs also match the true moral compass. If they do, it can only be because moral truth is identical to pragmatic need.
We have also clearly evolved a mental plasticity, which allows us, by way of culture, to overlay new behavioural tendencies, and we see this in the way moral mores change over time. We can teach our children to not only follow their instincts, but also to moderate and to some extent modify them. But again, there is no obvious reason why our cultural tendencies, which often reward either power interests, or the interests of social stability, should track some higher truth.
So, imagine a universe in which the beating of children is, in fact, morally good. How would the evolved brain, or the subsequent cultural overlay, ever discover this truth, let alone shape the brain to track it?
My argument is that we know of absolutely no mechanism by which this might happen and so, as with the levitating furniture, we should conclude it is not happening. (Some will argue that, unlike the furniture, we have clear evidence via our personal conviction, that moral truths exist. But ask yourself, should you meet a person with clear personal conviction, but no physical evidence, that the chairs levitate, would that also sway you?)
So, what’s the argument against this? Darrell suggests an answer can be found in this Stanford entry:
“Consider how a moral realist will approach the above debunking argument. It is intended to show us that realism is untenable, which of course means that this conclusion cannot just be assumed from the start. Yet if we begin the argument allowing that there may be independent moral truths, then why should we accept the initial claim about the pervasive influence of evolutionary forces on the content of our moral thinking in the first place? If there are independent moral truths, then we may plausibly have grasped many of them through autonomous exercises of our capacities for moral reflection, whereby we have come to recognize good reasons for thinking certain moral propositions to be true...”
The above makes two important claims. First, just because we may have established the way evolution can shape our moral intuitions, this does not in itself preclude the possibility that some of our moral intuitions refer to something more than mere selective contingency. In other words, science may provide one way of moral knowledge entering the system, but this need not be the only way. To assume in advance it is the only way is to beg the question.
The second claim made here is that, in fact, we do have another means of grasping moral positions, and this involves our capacity for reason, what here is referred to as autonomous reflection or autonomous moral reasoning. This is the observed human capacity to reason our way towards new truths. We have after all reasoned our way to the fact that there are infinite prime numbers, or that the world is round. Nobody claims our access to such truths is precluded by evolution. So, if we can reach mathematical or physical truths in this way, why not also moral truths?
The first point is valid, and one would be foolish to question it. An argument showing how evolution shapes moral judgement is not in itself sufficient to show either that there are no moral truths, nor that we have no access to them. The problem, it seems to me, comes about with the second claim. For, the argument I am making is that, while moral truths may exist (I have no opinion on this either way) we should conclude we have no access to them because we have no plausible mechanism by which they could be grasped. And that is an explicit claim that the appeal to autonomous moral reasoning does not work, if its intention is to show it to be a reliable generator of morally true statements. In other words, the analogy with scientific or mathematical reasoning is a false one. Here’s why:
Remember that the argument is not that we can not reason our way to moral beliefs, clearly we can and do, but rather that such reasoning will not reliably guide us towards external, or objective, moral truths. To see this, consider how both mathematical and scientific investigations work. Mathematics is essentially the process by which we unpack the implications of a set of axioms. So, in the case of prime numbers, we first set up a mathematical system, with a set of rules and definitions, and once these are accepted, we can work our way through various proofs (say the unlimited number of primes). So, we can say, given a particular set of axioms, a certain set of results must follow. The closest we can come to statements about mathematical truth is to say, if A is true, then it implies B must be true.
Clearly this is problematic if we hope that moral reasoning can reliably deliver up moral truths. For such reasoning to work, first we will need to define and accept our moral axioms. For example, a person starting off with the proposition that a moral goal is greatest benefit for the greatest number will be able to reason their way to a number of moral propositions that therefore follow (imperatives ruling tax systems, perhaps). However, the propositions thereby reached will only be reliably true if the starting moral assumptions are already true (that’s just how deductive logic works) and we must ask, well where did the truth of the starting assumptions come from? And, the evolutionist will argue, we know of no mechanism that would make the starting assumptions reliably true.
Science works slightly differently, in that both its starting assumptions and reasoned to propositions will then be tested against the world as we experience it. The belief that the world is round was tested against the angle of the sun at different geographic locations, for example. We say the model holds, not just because of the way we reason towards it, but also because of the way our experience of the world matches the model’s predictions. In moral reasoning, however, the real moral world exerts no influence against which we can test it (unless we redefine morality in pragmatic terms – the moral is that which fulfills us, for example – which was the evolutionist’s point in the first place).
So, where does all of this leave us? Well, we can still construct, consider and refine our moral systems, and set great store by them socially. We can use them to attain the goals we value, and to work towards a shared sense of humanity. What we can’t do, if we are to cleave to our scientific understandings, is hold that our moral discoveries are indicators of some higher, unchanging truth.