Haggerty, Imitation in Monkeys. 349 



ence of simple mechanical devices, the manipulation of which opened 

 doors, disclosed openings, or dropped food into the experiment cage. 

 The motives to action on the part of the monkeys were three: curi- 

 osity, the obtaining of food, and the tendencv to imitate. 



The problems which I have used are all comparatively simple. 

 It is an easy matter to construct devices which monkeys will not 

 manipulate, either on their own initiative or by imitation. The 

 results from such problems, however, have only a negative value in 

 the study of animal intelligence. To demand that an animal per- 

 form a wholly new act, that he behave in a way entirely diiferent 

 from his usual ways of acting, is a legitimate mode of procedure for 

 certain purposes. But if a monkey fails to manifest imitative be- 

 havior under complex and excessively strange conditions, it is not 

 proof that the animal lacks imitative ability. 



Human beings do not imitate all the acts of their fellows, not 

 even all those which it would be profital)le to copy, and to judge 

 by such failures would be to class man as a non-imitative animal. 

 This would be manifestly unfair, for in certain other situations 

 the imitative behavior will appear. The fact is that we imitate 

 most often in those situations in which wholly new elements are 

 few. We are reinforced by a great complex of habitual reactions, 

 and, when the new elements are mastered by imitation, these habitual 

 modes of activity complete the learning in a more or less auto- 

 matic way. Because we take advantage of our fund of habits is, 

 however, no reason to deny that our real advances in learning may 

 be by imitation. We do not demand that a person perform an act 

 wholly and entirely new before we credit him with imitative learn- 

 ing. 



We certainly should not be less generous with other animals. 

 They should be met as nearly as possible on their own ground and 

 presented with ])i-oblems in wliich they may have the advantage 

 of their fund of iidierited and acquired modes of behavior. At 

 first the elements entirely new should be as few as possible. If 

 they are then unaljle to ])rotit by seeing another animal perform 

 an act the case against their al)ility to learn by imitation would 

 seem to be conclusive. If under such simple conditions they do 



