THE PLAY OP ANIMALS. 179 



tative impulse as instinctive, and now I must return to 

 this vexed question. In order to get a clear view of 

 the opposite theory, according to which imitation is 

 of individual (not hereditary) origin, it is best to refer 

 at once to the work which more than any other has in- 

 fluenced modern association psychology — James Mill's 

 Analysis of the Phenomena of the Human Mind. So 

 far as I can see there is nothing essential in later eluci- 

 dations that is not contained in Chapter XXIV of this 

 book. Mill proceeds from the assumption that the 

 idea of a movement produces the impulse to perform 

 the movement itself. The motion of swallowing fur- 

 nishes a good example, for "if a friend assures you 

 that you can not refrain for the space of a minute from 

 this act, and you are tempted to try, you are almost 

 sure to fail." Why is this true? Because directing 

 the attention to the act of swallowing so strongly sug- 

 gests the muscular feeling attending the act, that swal- 

 lowing itself follows of its own accord. The same 

 result follows when an idea of motion is suggested by 

 the sight of it performed by another. For instance, 

 there are certain feelings which we hardly notice ac- 

 companying gaping, and when we see another person 

 gaping, we usually gape too; the act is so firmly asso- 

 ciated with the accompanying feeling that the sight 

 of the action arouses the feeling, which in turn calls 

 for the act in ourselves. This explanation is expected 

 to cover all phenomena included in the general name 

 of imitation. It will be seen that the awakening of 

 imitative impulse is here dependent on antecedent asso- 

 ciation. But for this it is necessary, as a rule, that the 

 act in question shall have been repeated frequently, 

 and thus, according to this theory, we would imitate 

 only such motions as are already familiar. Were the 



