child extends an idea obtained from the most impressive experience of childish difficulty—viz., "too big," so as to make it express the abstract notion "too difficult" in general.
In this extension of language by the child we may discern, along with this play of the feeling for similarity, the working of association. This is illustrated by the case of Darwin's grandchild, who, when just beginning to speak, used the common sign "quack" for duck, then extended it to water; following up this associative transference by a double process of generalization, using the sound so as to include all birds and insects on the one hand and all fluid substances on the other.[1] The transference of the name from the animal to the water is a striking example of the tendency of the young mind to view things which are presented together as belonging one to another and in a manner identical. Another curious instance is given by Prof. Minto, in which a child who applied the word "mambro" to her nurse went on to extend it by associative transference to the nurse's sewing machine, then by analogy applied it to a hand-organ in the street, then through an association of hand-organ with monkey to its India-rubber monkey, and so forth. Here we have a whole history of changes of word-meaning, illustrating in curious equal measure the play of assimilation and of association, and falling within a period of two years.[2]
There seems to be a like impulse to identify things which are closely conjoined in experience, as the extension of the word "spend" by the boy C so as to make it cover the idea of "costing." In like manner a child has been known to use "learn" for teach. In other cases we see a similar tendency to transfer a name from cause to effect, and vice versa. Thus, a little girl of four called her parasol when blown by the wind "windy," and the stone that made her hand sore a "very sore stone." In all these cases of transference it is evident that we have to do with two parts of a whole process, two aspects of one relation.
Here, again, one suspects the child is illustrating a common tendency in the growth of language. The etymological connection between the words teach and learn in German (lehren, lernen) shows that the human mind is apt to give a common name to closely related things. A west-of-England yokel still talks of "learning me"—i. e., "teaching me."
There is much, indeed, in the whole of these changes introduced by the child into our language which may remind one of the changes which go on in the growth of languages in commu-
