Re-evaluating Kant’s ‘Categorical Imperative’

Bayes’ theorem concerns how we formulate beliefs about the world when we encounter new data or information. The original presentation of Rev. Thomas Bayes’ work, ‘An Essay toward Solving a Problem in the Doctrine of Chances’, was given in 1763, after Bayes’ death, to the Royal Society, by Mr. Richard Price. In framing Bayes’ work, Price gave the example of a person who emerges into the world and sees the sun rise for the first time. At first, he does not know whether this is typical or unusual, or even a one-off event. However, each day that he sees the sun rise again, his confidence increases that it is a permanent feature of nature. Gradually, through a process of statistical inference, the probability he assigns to his prediction that the sun will rise again tomorrow approaches 100 per cent. The Bayesian viewpoint is that we learn about the universe and everything in it through approximation, getting closer and closer to the truth as we gather more evidence. The Bayesian view of the world thus sees rationality probabilistically.

As such, Bayes’ perspective on cause and effect can be contrasted with that of David Hume, the logic of whose argument on this issue is contained in ‘An Enquiry Concerning Human Understanding’. According to Hume, we cannot justify our assumptions about the future based on past experience unless there is a law that the future will always resemble the past. No such law exists. Therefore, we have no fundamentally rational support for believing in causation. Bayes instead applies and formalizes the laws of probability to the science of reason, to the issue of cause and effect.

I propose that we apply the same Bayesian perspective to Immanuel Kant’s duty-based ‘Categorical Imperative.’ This can be summarised in the form: ‘Act only according to that maxim which you could simultaneously will to be a universal law.’ On this basis, to lie or to break a promise doesn’t work as a practical imperative, because if everyone lied or broke their promises, then the very concept of telling the truth or keeping one’s promises would be turned on its head. A society that worked according to the universal principle of lying or promise-breaking would be unworkable. Kant thus argues that we have a perfect duty not to lie or break our promises, or indeed do anything else that we could not justify being turned into a universal law.

The problem with this approach in many eyes is that it is too restrictive. If a crazed gunman demands that you reveal which way his potential victim has fled, you must not lie to save him because this could not be universalisable as a rule of behaviour.

I propose that the application of a justification argument can solve the problem. This argument from justification is that you have no duty to respond to any request which is posed without reasonable appeal to duty. So, in this example, the gunman has no reasonable appeal to duty from you, so you can make an exception to the general rule.

Why is this consistent with the practical implications of Kant’s ‘universal law’ maxim? It’s an issue of probability. In the great majority of situations, you have no defence based on the argument from justification for lying or breaking a promise. So the universal expectation is that truth-telling and promise-keeping is overwhelmingly probable. The more often this turns out to be true in practice, the closer this approach converges on Kant’s absolute imperative by a process of simple Bayesian updating.

In a world in which ethics is indeed based on duty, it is this broader conception of duty which, I propose, should inform our actions.