Dennett's Philosophy: A Comprehensive Assessment Book

There are famously two sides to Kant's metaphysics. He is a 'transcendental idealist' in relation to the existence of the external world, but a representational realist in relation to the phenomenal world of our senses. Langton claims that humility derives from our inability to know given objects as they are in themselves, but she is also forceful in making a case that Kant is a scientific realist. This is in line with contemporary 'structural realist' interpretations of Kant, in which science furnishes us with explanations of the relations between objects rather than knowledge of the nature of the objects themselves. Thus we can remain neutral over the exact nature of unobservable scientific entities such as electrons and quarks while at the same time crediting them as 'real'.

For the scientific realist, mind-independent primary properties of objects such as extension, mass and charge are the business of science. Secondary qualities such as colour and texture are caused by primary properties and have no existence outside of the human mind. Langton believes that an opposing phenomenalist reading of Kant, in which primary and secondary qualities are on an equal footing and nothing but representations in us, is mistaken. What Kant actually thought is that the primary secondary distinction holds within the phenomenal realm. Phenomenal objects have the (often unobservable) properties that physics ascribes to them; and they do not have colours and tastes. Secondary properties should be identified with powers to produce the sensations in us, which they achieve through causation.

I am not enough of an expert on Kant to know what he really meant to say and it is probable that his views evolved over time. What I will say is that I share Berkeley's abhorrence of a scientific rationalism that denies reality to our rich sensible world of ripe strawberries, leaden skies and the smell of cut grass. We cannot conceive of objects without these secondary qualities. The essence of colour is that it is unquantifiable and not reducible to primary causes. Moreover, physical properties and the laws that relate them are abstractions from this manifest world. The law of gravitation states what happens to bodies upon which no other forces are acting; but there are no such bodies in the actual universe, and so strictly speaking it cannot be true of anything.