Although rarely made explicit, there are two predominant approaches to understanding the relationship between language and the world. One position holds that there are things in the world, and that words act as names for these things. These names, it is supposed, stand in for, grasp on to, or capture the things of the world. This is reflected in the way that some people try to teach language to children: ‘This is a ball, a cat, a dog’ – and they have pictures to back up their statements. Language is regarded as having a direct and fixed relation to the things of the world.

The second main approach is succinctly expressed in a line from Shakespeare: ‘that which we call a rose by any other name would smell as sweet’. Taken at face value, this line suggests that there are things in the world, such as roses, that have their own properties, such as smelling sweet. The words that we use to describe them, the names that we give them, may change or vary but the properties of these things will stay the same. In this sense, words do not map directly on to the world; they are just human artifices that are unable to fully capture the world. As the songwriter Paddy McAloon once put it: ‘Words are trains for moving past what really has no name.’

Both of these approaches have some force. Yet they also carry a range of problems. If words are taken as capturing the world, or indicating how the world is, the question arises of those words that can have different meanings on different occasions, thus appearing to pick out different items. It becomes difficult to maintain that words name things directly when one of the joys of language is that different words can be used to refer to the same thing. Cats and dogs are both animals and pets; a ball could be called a sphere or an orb. Moreover, one word can refer to two completely different things: a ball could be a dance. The second approach escapes such problems by emphasising the contingency of the meanings of words, and the lack of fixity in the relation of language to the world.