Language learning begins with listening. Children are greatly different in the amount of listening they do before they start speaking, and later starters are often long listeners. Most children will “obey” spoken instructions some time before they can speak, though the word “obey” is hardly accurate as a description of the eager and delighted cooperation that is usually shown by the children. Before they can speak, many children will also ask questions by gestures and by making questioning noises. Any attempt to study the development from the noises babies make to their first spoken words leads to considerable difficulties. It is agreed that they enjoy making noises, and that during the first few months one or two noises sort themselves as particularly expressive as delight , pain , friendliness, and so on. But since these can’t be said to show the baby’s intention to communicate, they can hardly be regarded as early forms of language. It is agreed, too, that from about three months they play with sounds for enjoyment, and that by six months they are able to add new words to their store. This self-imitation leads on to deliberate imitation of sounds made or words spoken to them by other people. The problem then arises as to the point at which one can say that these imitations can be considered as speech. It is a problem we need to get out teeth into. The meaning of a word depends on what a particular person means by it in a particular situation and it is clear that what a child means by a word will change as he gains more experience of the world. Thus the use at seven months of “mama” as a greeting for his mother cannot be dismissed as a meaningless sound simply because he also uses it at other times for his father, his dog, or anything else he likes. Playful and meaningless imitations of what other people say continues after the child has begun to speak for himself, I doubt, however whether anything is gained when parents take advantage of this ability in an attempt to teach new sounds.