Kant provides a simple example: a shopkeeper knows he can get away with
overcharging a customer, but feels moved to inform the customer that she has been
overcharged. So the ‘maxim’ is: ‘I should always be honest’ (Kant, 1996: 53). This
maxim can form the basis of a moral law only if it can be universalised, meaning
that anybody in the shopkeeper’s situation can make the same judgement, and the
shopkeeper in a different situation can apply that maxim. Universalisation entails
abstraction from people and situations. Perhaps the customer is a friend, and
friendship moves the shopkeeper to be honest, or alternatively, the customer is a
child, and the shopkeeper feels bad about cheating a child, or maybe the shopkeeper
‘just knows’ it is wrong to overcharge. These cannot justify the maxim because they
depend on the particular identities of the agents, or on particular emotions.
The categorical imperative is not a tool for making everyday judgements. This
becomes clear when Kant, in one of the formulations, maintains that one should
will that your maxim becomes a ‘universal law of nature’ (Kant, 1996: 73). This
indicates that the task is not to make case-by-case judgements but think ‘holistically’:
we imagine a societygoverned by universal laws. Such a society Kant describes as
a ‘Kingdom of Ends’, for if we universalise we must necessarily treat other human
beings as ends and not means (Kant, 1996: 80). In contrast to Locke, these laws
are not given to us by God, or through our senses, but are ‘constructed’ by human
beings exercising powers of reason. Through construction of moral laws we lift
ourselves above our animal natures and prove our autonomy. There is a crucial
political point here: we can be coerced into conformingwith what morality requires,
but we cannot be coerced into acting for the right reasons. The shopkeeper can be
motivated to be ‘honest’ by threat of punishment, but he would not be acting morally
because he is not being moved by reason.
Some contemporary political theorists draw an anarchist conclusion from Kant’s
argument. Robert Paul Wolff argues that we can never reconcile moral autonomy
and political authority (Wolff, 1970: 18–19). But, in fact, in his political writings
Kant does defend the state. He even maintains that a civilised state is possible among
a ‘nation of devils... just so long as they get the constitution right’ (Kant, 1996:
335). To understand the relationship between morality and politics we need to
distinguish internal freedom and external freedom. The former – which can also be
called autonomy – entails the ability to be motivated to act morally by the force of
reason alone. The latter is the idea that the freedom of one person must coexist
with the freedom of all others. This is expressed as a system of rights, coercively
enforced by the state.
The state serves the end of morality by helping to realise the ‘Kingdom of Ends’.
The difficulty with this argument is that human agents must will the creation of
that ‘Kingdom’, whereas in a political community – under the state – we are coerced
into behaving in accordance with other people’s rights. Attempting to resolve the
conflict between autonomy and coercion has been central to the liberal project. One
way of resolving it would be to posit two standpoints that a citizen can adopt: the
standpoint of moral autonomy and the standpoint of a subject of law. As an
autonomous agent you will the creation of a political community in which each
person’s rights are respected, but you also know enough about human nature to
recognise that rights will have to be protected through coercion, such that you are
at the same time willing the creation of a coercivepolitical community. This would,
Chapter 8 Liberalism 185