The purpose of this study is to make clear the function of family toward offense through “delinquents” and “non-delinquents” families. The subjects for the present study consisted of 83 families which more than a half of sibling committed offenses and 20 families which no members committed offenses though they were reared in home broken through desertion, separation, alcoholic of one or both parents, or financial poverty.

The main findings of this study were:

1) Among delinquents there was a greater tendency for their fathers to appear clearly as drunkenness, ill-health, tyranny, debauchery, and criminal than among non-delinquents.

2) Among delinquents their mothers, when in trouble, tended to desert their homes.

3)Most of paternal misbehaviors were found in themselves before they got marries, whereas most of maternal misbehaviors were produced as the result of abnormal behavior on the part of fathers.

4) When family finances was broken through paternal misbehavior, mothers tended to run away from their husbands and children.

5) Among delinquents they came largely from homes which were economic misery, and most of them lived in the slums,

6) Among delinquents the criminality rate of brothers was more than three times the number of that of sisters, and 35% of brothers without sisters committed offenses (87% of brothers and 29% of sisters committed offenses), whereas among non-delinquents there was no comparable difference between brothers and sisters.

7) There was a greater tendency among father-absent families to seek members of the family for their partner in the crime than among mother absent families.

8) There was a few cases suggested that if they had fathers and forefathers with them as the model for identification, although they were reared in broken home, they should be prevented from committing offenses.

9) Among non-delinquents they tended to evaluate an event in relation to productive desires, whereas among delinquents they tended to evaluate an event in connection with attitudes such as ill feeling, idleness.

The problem of the interrelationship of the family as a social system to juvenile delinquency as a social phenomenon can be rightly understood only when we, students, examine precisely the way the family relates to each action space of delinquency. However closely each factor of the family or personality traits isolated from the action space may be related to the phenomenon, the problem cannot be solved either synthetically or scientifically.

The writers, in this paper, intended to understand historically the interrelationship of the family to the whole “delinquent formation space” in juvenile delinquency, which consists of the formation and fixation situation of delinquent readiness, the occurrence situation of delinquent behavior, the control or suppression situation of delinquent behavior, and the correction or dissolution situation of delinquent readiness.

And the relationship of the family to the delinquency especially in middle adolescence was divided into two major types on the basis of the point that whether or not the conditions in childhood and in early adolescence functioned as delinquent readiness for delinquent behavior in middle adolescence. Each type was devided, moreover, into two subtypes according to the point that whether or not the occurrence situation of delinquent behavior was formed in the earlier period of development.

In each of these types, the family had different relations to the formation and fixation situation of delinquent readiness, to the occurrence situation of delinquent behavior, and to the suppression situation of delinquent behavior, though the relationship was fundamentally indirect because middle adolescence corresponds to SD-II or -III. And in every type the family is often requested to relate itself to the correction situation of delinquent readiness by the public organizations. And this relationship plays an important role in the later delinquent behavior. For each type the relationship was closely examined in connection with Delinquent-Geno types.

This typology is applicable to the delinquency in both early and late adolescence, though there is some difference in the degree of strength of the relation among the three stages of adolescence. In the case of early adolescence, the family has direct relation to the formation and fixation situation of delinquent readiness and the occurrence situation of delinquent behavior. For example, the victim, or the cooperator, or the protector of the delinquent behavior is the family or family members, and sometimes delinquent behavior place is in the family. And since early adolescent bases his essential ego on the family, the family does necessarily have great importance in the other situations of delinquent formation space, too. And in the case of late adolescence, the family, which he will form in adulthood, comes to have important meaning in each situation of delinquent formation space.

In short, the relationship of the family to the delinquency should not be considered as the same structure throughout one’s life cycle. And the difference in Socialization-Depth of individuals determined primarily the interrelationship of the family to delinquency or crime.

In criminological practice, it has been needed to develop some sort of screening tests to differenciate personality disorders or psychopath. We, therefore, attempted to compare the results of two psychological tests in psychopathic delinquents group (PD) with those in non-personality-disorder delinquents group (ND), and to establish objective and practical standards for differenciating the psychopath in connection with the treatment.

The multiple correlation ratio method was applied to analizing 13 scale scores of the Ministry of Justice Personality Inventory (MJPI) and 11 categories relevant to psychopathic syndromes from the Ministry of Justice Sentence Completion Test: Form Youth-I (SCT). According to this method, structure coefficients and weight coefficients were calculated. The psychopathic scale scores for each of subjects were also calculated by using these weight coefficients.

126 subjects (PD:88, ND:38) for the MJPI, and 131 subjects (PD:49, ND:82) for the SCT were selected from among the sample of the previous study, “A Study on the Diagnosis of Psychopathic Juvenile Delinquents.” (1973)

Among structure coefficients of 13 scale scores of the MJPI, those of Paranoid (P), Depression (D), Deviation (De), Unstable (U), and Explosion (X) were found to be significant. Throughout some discussions on psychological meanings of these 5 scales, it was concluded that these were at least related to emotional unstability.

Weight coefficients of P and D were also found to be higher than those of another scales. Although weight coefficients were not always concerned directly with the psychopathic scale, it could be taken for granted that P and D, representative of paranoid trends, anxiety trends respectively, were fundamental factors in differenciating the psychopath.

The means and standard deviations of the psychopathic scale scores with the MJPI were 0.43, 0.82 for PD, and -0.99, 0.58 for ND, respectively. The multiple correlation ratio was so high (r=0.65) that the psychopathic scale might be put to practical use.

As for the SCT, the categories indicating high plus structure coefficients were “lack of emotionality and/or bad human relations”, “lack of durability”, “unstability or immaturity of emotion” and “ignorance of social norms and/or group norms”. On the other hand, the categories indicating minus structure coefficients were “persistency” and “erasing and/or omitting letters”. Moreover, the response frequencies in these categories were found to be almost same for both groups.

In most categories, weight coefficients ran parallel to structure coefficients. But, “lack of guilt feelings and/or inability to profit by or to use experience” showed minus score in weight coefficient, although it showed plus score in structure coefficient.

The means and standard deviations of the psychopathic scale scores with the SCT were 0.85, 0.83 for PD, and -0.51, 0,69 for ND, respectively. In addition, the multiple correlation ratio was fairly high (r=0.61). These results indicated that there was a significant difference between PD and ND, as it was showen for the MJPI.

The present study was intended to investigate the characteristics of the moral judgement of juvenile delinquents, standing on the view-point that the judgement of rightness or wrongness would suggest the permissiveness for transgression. The story-completion test has been widely used to examine the problem of moral judgement, but we employed the questionnaire method as the numerical method to measure the severity of moral standards.

The subjects were 100 normal high school boys and 70 male delinquents living at the juvenile training school.

The questionnaire used in this study consisted of 40 morally prohibited activities. Each item offered a five-point scale for rating and each subject was asked to rate the degree of wrongness of these items. Spearman-Brown’s coefficient of reliability was computed for the scale of 38 items shortened by an item analysis, and found to be 0.931 for the normal group and 0.954 for the delinquent group.

It is doubtful to regard the delinquent group as a homegeneous unit, since the delinquency is the legal classification. Therefore, both groups were devided to two sub-troups on the basis of E scale scores of Mausely Personality Inventory respectively, and 30 subjects were extracted from each subgroup to analyze the data.

First, 38 items were classified to five categories (illegal conduct, pre-delinquency, transgression of public morality, violation of traffic law, and transgression of personal standard) and the separate simple analysis of variance was performed for each category. In all categories the group factor was statistically significant, but the personality factor and interaction effect were not. It was found that the moral judgement of delinquent boys was laxer than one of the normal boys.

Secondly, the group differences were separately tested by a chisquare analysis for each item, and the significant differences were found in about 60% of the items. When the items showed significant differences between the groups were compared with the items showed no significant differences, it was found that the activities of the former items were relatively wronger, but that them of the latter were less wrong.

Finaly, the factor analysis (principal axis method) was performed to classify the 40 items to some categories. The product-moment intercorrelations among 40 items were calculated for normal group, and 9 factors were extracted from a correlation matrix. Then the group differences were tested by t-test for the cluster of the items related to each factor axis, seperately. As a result, it was found that the group differences were significant in the middle or upper class of moral value hierarchy, but that not in the lower class.

The conclusions from these findings are as follows:

(1) The moral judgement of juvenile delinquents is totally laxer than one of the normal boys.

(2) This tendency is not found in all prohibited activities but found in the relatively wronger activities.

(3) It is interpreted that this fact is due to the undifferentiatedness of the moral value system of juvenile delinquents.

Two experiments using verbal conditioning were conducted to test the assumption that High Pd delinquent. In Exp. 1, the Ss were required to combine different verbs with one of five different personal pronouns: “I”, “YOU”, “HE”, “SHE”, “THEY”. The Ss of reward condition were reinforced by the experimenter with a mild affirmatory word (“good”) at the end of the sentence the Ss constructed and the Ss of punishment condition were reinforced with a mild affirmatory word (“bad”), each time they used a pronoun which was of the highest operant level in the operant conditioning phase. The findings showed that in reward condition the performance of the conditioning trials decreased as a function of Pd score: On the other hand, in punishment condition, the higher the Pd score, the more the performance increments.

The task used in Exp. 2 was similar to one in Exp. 1. The Ss were divided into four patterns from the combination of both Pd score and heart rate level. The delinquent of the pattern of High Pd and low heart rate showed a significant performance gains. However, successful conditioning was not found in the Ss of other three patterns.

These results failed to support the assumption that learning did not occur in High Pd delinquents because of their diminished state of physiological arousal, but suggested that successful conditioning in High Pd delinquents depended upon the stimulus situations.