The stimuli that arouse movements of escape are of two sorts: those that directly cause some irritating sensation, and those that are simply signs of danger. The smaller avoiding reactions--flexion reflex, coughing, etc.--are {143} aroused by stimuli that are directly painful or irritating; whereas flight, cowering, etc., are mostly responses to mere signs of danger. A "sign of danger" is usually seen or heard at some distance, not felt directly on or in the body. Now, while avoiding reactions are attached by nature to the irritating stimuli, it is not at all clear whether escape movements are _natively_ attached to any signs of danger, or, if they are, to what particular signs of danger they are attached. What visual or auditory stimuli, that are not directly irritating, will arouse escape movements in a young child? For the youngest children, no such stimuli have been found. You can easily get avoiding reactions from a little baby by producing pain or discomfort; you can get the clinging response by letting the child slip when he is being held in your arms; and you get crying and shrinking on application of a loud, grating noise, such a noise as is irritating in itself without regard to what it may signify. But you cannot get any shrinking from stimuli that are not directly irritating. For example, you get no sign of fear from a little child on suddenly confronting him with a furry animal. With older children, you do get shrinking from animals, but it is impossible to be sure that the older child has not _learned_ to be afraid of them. I have seen a child of two years simply laugh when a large, strange dog came bounding towards him in the park; but a year later he would shrink from a strange dog. Whence the change? There are two possibilities: either a native connection between this stimulus and the shrinking response only reached its maturity when the child was about three years old--and there is nothing improbable in this--or else the child, though actually never bitten by a dog, had been warned against dogs by his elders or had observed his elders shrinking from dogs. Children do pick up fears in this way; for example, children who are {144} naturally not the least bit afraid of thunder and lightning may acquire a fear of them from adults who show fear during a thunderstorm. On the whole, the danger-avoiding reactions are probably not linked by nature to any special signs of danger. While the emotion of fear, the escape impulse, and many of the escape movements are native, the attachment of these responses to specific stimuli--aside from directly irritating stimuli--is acquired. Fear we do not learn, but we learn what to fear. Crying.