Criminal behavior seems to develop in the following way: Among two- and three-year-olds getting their first taste of the belt, the reaction appears to be one of confusion and bewilderment. The child's adrenalin level is kept high with frequent punishments and by the time he enters school, he is literally flying around the room and is seen as hyperactive or as "emotionally immature." The school now calls the parent, repeatedly saying that the child is a behavior problem, and the child gets worse beatings than before. While the child may occasionally run away or rebel in some way, there appears to be a latency period during which the child becomes resigned to his fate. He may react in such passive-aggressive ways as refusing to do his school work, wetting the bed, or being a general nuisance, but delinquency does not seem to rear its ugly head until the child is 11, 12 or 13.

Most parents of delinquents tell me that they had no problems with their child until he was 12 or 13. Not coincidentally, parents seem to stop beating their children at about the age of 13 or when the child says, "If you hit me once more, I'll let you have it."

Unusually high adrenalin levels are found in sociopaths in prison, yet they do not seem to respond appropriately to adrenalin output. They seem to have learned not to attend to fear responses. In laboratory tests of learning they did very poorly under punishment conditions, although they did as well as normals with a reward contingency. Were they constitutionally defective? Or was their early training responsible?

Among court-referred boys and welfare referrals the degree of aggressive behavior correlates with the severity of punishment they have received.

Dr. Welsh is a clinical psychologist practicing in Bridgeport, Connecticut. A major part of his professional activity over the years has been his work with court adjudicated youth and their parents.