This article follows Once more without feeling.
It continues a discussion which started in The ethics of belief.
We are now getting into the meat of the categorical imperative itself.
This will not be an exhaustive or comprehensive account. The focus will be on what kind of thing the categorical imperative is – its logical status.
Kant offers a number of different formulations, of which the first is:
I ought never to act in such a way that I couldn’t also will that the maxim on which I act should be a universal law.
[Immanuel Kant, Groundwork for the Metaphysic of Morals1]
Perhaps the most successful example he gives is that of keeping promises out of duty:
Consider the question: May I when in difficulties make a promise that I intend not to keep? …
…How can I know whether a deceitful promise is consistent with duty? The shortest way to go about finding out is also the surest. It is to ask myself:
Would I be content for my maxim (of getting out of a difficulty through a false promise) to hold as a universal law, for myself as well as for others?
Could I say to myself that anyone may make a false promise when he is in a difficulty that he can’t get out of in any other way?
Immediately I realize that I could will the lie but not a universal law to lie; for such a law would result in there being no promises at all, because it would be futile to offer stories about my future conduct to people who wouldn’t believe me; or if they carelessly did believe me and were taken in, would pay me back in my own coin. Thus my maxim would necessarily destroy itself as soon as it was made a universal law.
In this example (as also in the more general context of lying) the imperative is revealed as a logical one, once the universalising premise is added. If no one kept their promises, the institution of promising would not survive, or would not even exist. So it is not this:
Promising is a good thing
Ergo anything which endangers it as an institution is wrong
Ergo it is one’s duty to protect and preserve the institution
The morality is contained in the principle of universality itself. It is wrong to do something you would not voluntarily universalise.
The reason for this is to do with rationality and freedom. Rationality is to behave according to laws. Laws are by definition universal. Freedom is autonomy. So to be both free and rational one must behave in accordance with laws, but only laws one sets oneself. Laws must apply universally, therefore the laws one sets oneself must apply universally.
‘Do not murder’ should be a similarly straightforward example. If I decide to kill someone I do not like I must also will that anyone else can kill people they do not like – which could include me. But it does not make sense for me to make a law which would allow someone else to kill me.
The content of any particular law has no bearing on the formulation of the categorical imperative, as this content comes afterwards. Specific imperatives like ‘do not lie’, ‘keep your promises’, and ‘do not murder’ are binding because the formulation can be demonstrated to apply to them.
As the name suggests the categorical imperative is categorical. This is to distinguish it from a hypothetical imperative, eg of the form:
(i) If you want to (or have to) achieve y, then do x.
A categorical imperative on the other hand is simply of the form:
(ii) Do x.
This does not necessarily mean that any statement containing an ‘if’ clause cannot be a categorical imperative. Consider for example:
(iii) If you are male, do x.
In the hypothetical imperative (i) x is a means to y, where y is an end. In the categorical imperative (ii) x is an end in itself.
In (iii) however the ‘if’ clause is not an end, but a qualification. Whether (iii) can be said to hold universally is ultimately a question of semantics. It holds universally in the sense that everyone should ask himself or herself ‘Am I male?’ and if the answer is yes then he should do x.
It does not seem wildly different from the example of promising:
(iv) If you have made a promise, keep it.
The only apparent difference is that the person typically chose to make the promise, but did not choose to be male. Is this significant? Consider:
(v) If you are an employer, do x.
Typically a person would choose to be an employer. Imperative (v) holds universally in the sense that everyone should ask himself or herself ‘Am I an employer?’ and if the answer is yes then he or she should do x.
But it is not in fact obvious how relevant deliberate choice is. Kant is careful to emphasise that the categorical imperative does not just apply to all humans, but to all rational beings – regardless of whether there happen to be any rational beings other than humans. This is important because he wants to claim the categorical imperative is binding on all rational beings because they are rational, not that it is binding on all humans because they are human.
The snag is that the category of ‘possible rational being’ could cover all sorts of entities. It could include, at one extreme, immortal and/or omniscient and/or omnipotent beings; at the other extreme, rational beings who may be like humans in some ways but are in other ways very differently constructed and/or wired; and somewhere in the middle, any combination of features.
Take a far-fetched but logically possible example of a community of beings who are incredibly perceptive, so much so that they can read each others’ minds so accurately that they are nearly always right. But they are also almost unbearably sensitive, and suffer horribly when they are forced to think about things they would rather not think about. In such a community the imperative ‘do not lie’ would be of questionable value, because everyone would know when someone was lying. Telling the truth, on the other hand, could lead to avoidable suffering.
If I was one of those hypersensitive souls, could I say to myself ‘I will always tell the truth’ and at the same time will that it was a universal law, so that everybody else would also always tell the truth, regardless of the unnecessary suffering it caused, both to myself and others? Unlikely.
So if we now imagine a community of humans side by side with a community of hypersensitive souls, the specific imperative about telling the truth would need to be worded differently for each community. The gist might be something like:
(vi) If you are a human, tell the truth; but if you are a hypersensitive soul, say what will cause least suffering.
The problem this causes is that if a specific imperative (ie a willed universal law) can be structured like this, with qualifications, then surely something like this could also ‘pass’:
(vii) If you are a white male, take priority.
The fact that (vi) is ‘moral’ and (vii) is ‘immoral’ is unfortunately beside the point, since we are trying to establish the categorical imperative to be the foundation of morality.
© Chris Lawrence 2008.