up “f \“ ‘\ 5‘! all“ Elf“ b? \ ¢ \ i[WLNWln/‘IEI‘W«J 122 203 A PERSONALETY 5WDY OF TWO TYPES OF MURDERERS Thesis for 1'5"; Degree of DH. D. MICHIGM STATE UNIVERSITY Miies Pothast 1956 This is to certify that the thesis entitled A.Poraona11ty Study of Two Types of p Murderers presented bg Miles D. Pothast has been accepted towards fulfillment of the requirements for 62 . IQ degree in 6% way? / $7. é'fiaf—“~ ('7 Major professor Q‘lAA’KJ‘ / 3/ /9:‘>‘_(;9 Date O~169 A PERSONALITY STUDY OF TWO TYPES OF MURDERERS By 09“] MILESlPOTHAST A THESIS Submitted to the school of Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Philosophy \\3 \\ ~ L \\ \l 1956 ‘ K \ 5/02 3/57 fixae/ ABSTRACT An eXploratory personality study of forty convicted and imprisoned murderers was conducted by the use of the psycho- logical tools of the Rorschach Test and the Minnesota Multi- phasic Personality Inventory. These forty murderers consist-‘ ed of a group of twenty men who had been convicted of murder- ing their wives (this group was designated as “Passion“ mur- derers) and a group of twenty men who had been convicted of murdering while committing armed robbery (this group was de- signated as 'Profit' murderers). These two murderer groups were equated closely as to age, I.Q., socio-economic background, color, and time spent in pri- son on present sentence. The results of their performance on the Rorschach Test and the M.M.P.I. was compared.with each other, a prisoner group and a non-prisoner group. Obtained results showed both murderer groups' performance to deviate considerably from that of the non-prisoner groups, to a much lesser extent from that of the prisoner groups and even in some aspects from each other, with the "Profit" mur- derer group being more deviant than the "Passion“ murder group. Analysis of the findings of these M.M.P.I. and the Rorschach findings reveal the "Passion' murderers to be ex- tremely immature,emotionally labile, easily disturbed and un- able to form good interpersonal relationships, to utilize ii strongly repressive neurotic type defenses, and to show a marked tendency toward antisocial behavior. The “Profit" murderers were found to be extremely immature psychologically and emotionally labile but did show a lesser tendency for emotional reaction than did the "Passion" group. Further- more, the “Profit'_group was found not only to be unable to form good interpersonal relationships, but to avoid close contact with others; showing a social as well as strong antisocial tendencies. The obtained results lead to the implication of the presence of psychopathic personality types such as have been proposed by Gilbert (36), with the “Passion“ murderer resem- bling the "hysterical psychOpath" and the “Profit” murderer the “schizoid psychopath.‘ Further evidence is needed in this area and it was suggested that analysis of similar data obtained from convicted murderers in other regions of this country would be helpful in evaluating the personality structures of murderers and that data from the members of other criminal categories, such as armed robbery, be evalu- ated and compared.with that of murderers. 111 ACKNOWLEDGEMENTS The author wishes to express his sincere thanks and appreciation to Dr. G. M. Gilbert, whose contributions to this investigation included not only most helpful super- vision, but inspiration and unfailing interest as well. ‘The results of this investigation are herewith dedicated to Dr. Gilbert. The author is further indebted to Dr. Charles Hanley for his invaluable advice and interest; to Mr. Ray Butchkoe and others too numerous to mention for their assistance in obtaining the data, and to his wife, Donna, for her en- couragement and assistance throughout. iv 11. III. 1V. VI. VII. VIII. 1X. TABLE OF CONTENTS INTRODUCT ION O O O O O C A. Historical Background . . B. Review of Survey Studies 1. Primarily sociological studies 3. Psychologically-oriented PROBLEM . . ... . . . . . . METHODOLOGY . . . . . . . . A. Sample . . . . . . . B. Test Materials and Procedure RESULTS 0 O O O O O O 0 0 O A. Rorschach Findings . B; Minnesota Multiphasic Test Findings . . . DISCUSS ION o O O O O O O O O A. Rorschach Test . . . B. Minnesota Multiphasic Inventory . . . . . . CONCLUSIONS . . . . . . . . SUMMARY . . . . . . . . . . APPENDICES . . . . . . . . BIBLIOGRAPHY . . . . . . . Personality Personality studies Page 36 42 43 49 56 58 60 65 LIST OF TABLES TABLE Page I. Homicide Rate by Countries, Per Hundred Thousand of Population . . . . . . . . . 1 II. Mental Diagnoses (Psychiatric) of 781 Murderers . . . . ... . . . . . . . . 11 Ill. Age and I.Q. of Two Groups of Convicted Murderers. Age of 80 Youngest of Rapa- port's “Well-Adjusted Patrolman" . . . 32 IV. Mean Scores on the Rorschach Test for the Two Murderer Groups, Rapaport's 'Well—Adjusted' Patrolmen Group and Lindner's “Normal“ Prisoner Group . . . OJ C»- V. Probability Values Obtained by Use of the Kolomogorov-Smirnov Method of Inter- group Comparisons on the Two Murderer Groups and Rapaport's “Well-Adjusted" Patrolmen' Group . . . . . . . . . . . . 34 VI. Mean Scores on the M.M.P.I. for Two Murderer Groups; Both Individually and Combined, a “General Prison" Group, and Schmidt's “Normal“ Group . . . . . . . . 35 VII.a M.M.P.I. Profiles for the “Combined" Murderer Group, the General Prisoner Group and Schmidt's “Normal" Group . . . 33 V11.b M.M.P.I. Profiles for the 'Profit' and the 'Passion' Murderer Groups . . . . . 37 VIII. Mean Differences and T-Values for Differences on M.M.P.I. Scales Between the Two Murderer Groups, a General Prisoner Group and Schmidt's Normative Gr CUP 0 e e e e e e e e e e e e e e 0 fi IX. Mean Differences and T-Values for Differences on M.M.P.I. Scales Between the Combines Murderer Groups, the Gen- eral Prisoner Group and Schmidt's Nor- mative Group . . . . . . . . . . . . . . 56 vi TABLE XI. LIST OF TABLES - continued Page M.M.P.I. Raw Score Means and S.D.'s for the Two Murderer Groups, Both as Sepa- rate Groups and as a Combined Group, and the General Prisoner Group . . . . 40 Mean Differences, S.E. of Mean Differ- ences and T-Values for Differences on M.M.P.I. Raw Scores Between the Two Murderer Groups, Both as Separate Groups and as a Combined Group, and a General Prisoner Group . . . . . . . 41 vii 1. INTRODUCTION The United States has long been known as one of the most murderous countries in the world. That this is fact and not merely a legend due to or enhanced by the sensa- tionalism and widespread publicity of the "Chicago gang- ster' era of the 30's, can be seen in the comparative statistics of Table I ( 33 ), ( 44 ), ( 53 ). Table I HOMICIDE RATE BY COUNTRIES PER HUNDRED THOUSAND OF POPULATION Country Years Rate Years Rate United States 1911-1931 7.3 1948-1951 5.30 Italy 1910-1920 3.60 Canada 1914-1930 3.8 l947-1950 2.9 Australia 1911-1920 1.90 South Africa 1913-1918 1.80 Spain 1911—1917 0.90 Sweden 1914-1916 0.72 1939-1942 0.47 England and Wales 1910-1919 0.43 1948-1951 0.35 The Netherlands 1911-1920 0.13 - 1921-1927 0.06 According to the Metropolitan Life Insurance Com- pany ( 53 ) homicide is presently on the decline in this country, with the national rate being 5.8 deaths per 1 100,000 persons in 1934 but only 1.9 per 100,000 in 1954. Even at this relatively “low“ rate, this means that about 3,000 people were murdered in this country in 1954. The U.S. homicide rate is still six times that of England, Wales, and Ireland and three times that of Scotland. It is true that there are wide regional and reacial differences in homicide rates within this country, with the South's negro population having a disproportionately high rate, but this is only part- ly accountable. As for ours being a young, expanding country with a “mixed frontier“, American whites are being murdered at more than twice the rate of Australians and Canadians, where the frontiers are far more primitive. At the present time, there are probably more convicted murderers in the prisons and institutions in this country than there ever have been in the past. This is largely due to the fact that the death penalty is steadily losing ground and has been completely discarded in many states -- without any corre- sponding rise in the homicide rate. Several states, including Michigan, are even paroling some of these individuals who for- merly would have been executed. The question of just what sort of peeple murderers are and whether they can be trusted not to repeat such an offense thus gains added social impor- tance. It is the purpose of this study to investigate the per- sonality characteristics of a selected sample of murderers by the use of standard psychological tests. This is of necessi~ ty a rather exploratory study, as a search of the literature revealed no previous study of this type on such a population. A. Historical Backgrgund It seems that very little consideration was given to the prevention of death by homicide until the influence of mental hygiene made itself felt around the turn of the present cen- tury. Garofale ( 33 ), one of the great sociOIOgiats of his time, in 1885 wrote that he considered the “typical criminal“ to be “morally a monster possessed of some characteristics in common with the savages and still others which sink him be- neath the level of humanity“ and the murderer to be at the summit of criminality. He did, however, admit the existence of a class of what he termed to be “negligent murderers“ in whom he felt moral anomaly was not necessarily present. Far- ri ( 19 ), a contemporary of Garofalo, postulated the exis- tence of five categories of criminals; including the murderer in the grouping of “criminal by passion“. This group he be- lieved to have an excess of the trait he designated as “pas— sion“ and described them as persons of a “sanguine“ and “ner- vous“ temperament who could be easily distinguished by their anthropolOgical traits. He said they had.an exaggerated sane sitivity which is quite the reverse of the born habitual crim- inal and he believed that they were somewhat related to the insane and the epileptic, and further stated that these ten- dencies may merely be a concealed epileptic tendency. Smith (.53 ), was one of the first, if not the very first, of mo- 4 aern criminologists to recognize the importance of modern dy- namic psychology in helping to understand the murderer. He suggested that homicidal impulses are due to a conflict in early life which has given rise to a repression (which he felt was probably always of a sexual character) and produced the mental attitude of doubt and uncertainty. The emotional energy of this repressed complex has become detached and trans- ferred to something in consciousness but the obsession to which it is transferred represents a very great distortion of the repressed complex, just as one gets distortion in dreams. Brasol ( 9 ), subsequently advanced the opinion that murder- ers are not necessarily insane and that not every homicide is necessarily an act of compulsion. In 1933, White ( 5? ), of~ fared the interpretation that murder is the result of an ex- plosion of a man due to aggravation of his aggressive tenden- cies as a result of anger or acute emotional disturbance. The real motives thus could not be understood until the deep mech- anisms were analyzed. According to this explanation, some apparently deliberate cases of murder really are accidental as the result of a chain of circumstances impinging on an in- dividual who had no overt desire to commit such an offense. As Selling ( 52 ), points out, White doesn't extend his dis- cussion this far, but it would be necessary that even such a man have had a background - expressed in terms of passion, fear, or circumstances in his external environment which would enable him to commit an offense - or he would not have resorted tc homicide to solve his problem. From the strict- ly psychoanalytic point of view Raik ( 49 ), has gone into great detail in discussing the fact that the thinking pro- cesses of the murderer can be carefully analyzed and a know- ledge of his past life can be correlated with the type of crime so as to indicate certain mental processes. He stresses particularly here the importance of Totem and Taboo and.magic thinking. The ”father—surrogate” eXplanation of murder, a mechanism now commonly enacted by the psychoanalytic school, was given expression in the writings of Alexander and Staub ( l ), in 1931. This point of view holds that the mur- derer“s superego won't allow him to satisfy his hatred for the father by killing him. Alexander and Staub, as well as other modern analysts, also believe that the Oedipus complex plays the central role in the development of “perversions“ and that murder is often the representation of parvertive tendencies. Selling ( 52‘), goes along with the Adlerian school of interperative psychology and points out that the basically neurotic individual may kill to escape from some mental problem - for example in order to compensate for basic feelings of inferiority he may kill some individual who he feels is superior to him. Banay ( 3 )5 feels that murder knows no distinctions, that anyone may be a potential murderer under the right cir- cumstances. In his opinion, a scientific evaluation and classification of the murder of one human being by another can be made only in broad lines, beyound which a true under- standing of such a person can only be achieved in individual terms and reactions. In his words “the psychological profile of the individual murderer varies as greatly as does his phys- ical profile and the forces that operate the personality are so varied and variable that a complete understanding of the action can be evaluated only by the study of the individual in his total functioning.“ Karpman ( 35 ), takes the point of view that murder is a highly complicated form of behavior and must be viewed as the culmination of life-long difficulties which may well have had their origin in the early life of the individual. He feels that this even extends to the victim - that there is a highly specific relation between the murderer and his victim, often with symbolic value. Research into the area of homicide has expanded in re- cent years, especially in investigation of the sociological causes and concomitants. The majority of the research in this area that has been published by psychologists and/or psychiatrists, however, has been of the intensive single case study type (often as not that of unique and highly pub- licized cases such as that of Heirens, LeOpold and Loeb, etc.) which do not have too much to offer scientifically in terms of the personality characteristics of those who com- mit murder. Gilbert's ( 86 ), study of the Nazi Gestapo mass murder machine uses the single case study as an illus- trative point of departure for a more sophisticated analysis of social, institutional, interpersonal and pathological in- terrelationships in this more complex phenomenon. B. Review 9f Survey Studigs l. Primarily sociological studies The high rate of homicide in the United States has a- roused considerable concern among the sociologists of this country and has come to be regarded by some students of pubs lic health (most notably Whitfield ( ss ), and Selling ( 53 i as one of the foremost public health problems in the United States. Whitfield ( 58 ), describes the causes of homicide rather vaguely as: (1) predisposing, such as heredity, envi- ronment and illiteracy, and (3) active, comprised of “rob- bery, revenge, jealousy, gambling, drunkenness and immorality? He states that the most important cause of homicide is the lack of law enforcement. He cites as proof of this the fact that the murder death rate in England is much lower than that throughout the U. 8. A., disregarding the obvious sociological and psychological differences between the two countries. Selling ( 52 ), in a study of 242 murderers, found that they tended to be predominantly intellectually inferior males, practically all of whom were insane. He considers the racial factor to be a significant influence in homicide because of the “inferiority feelings arising from being a member of a minority race“. Brearley (lO)-(ll), a sociologist, regards the prevalence of homicide as a prime indicator of man's failure to achieve “social well-being“ and of the inadequacy of social control. He feels that homicide and other types of anti-social behavior must receive greater attention if human welfare is to keep pace with material progress. In his survey of homicide in the United States he concluded that in any attempt to understand why the homicide rates in the united States are so much higher than for other civil- ized countries, emphasis should be laid upon the presence of influential folkways and cultural patterns in this country, most of which are survivals of more barbarous days when hu- man life was little esteemed. He felt this to be especially true in the case of colored people, who he found to have a homicide rate almost 7 times as great as that of whites, pointing out that in Central Africa the sacredness of human life is not greatly emphasized and that this trait was brought to the United States by the Negro slaves, and was subsequently strengthened by the treatment they received here. Brearley does admit, however, that the differences in homicide rate between Negroes and whites may be more appar- ent than real and might tend to disappear if educational and sociolOgical factors were held constant. Virtually all of the studies of homicide reviewed by this writer (exclusive of the single-case studies) offered rather similar socio- economic and/or cultural explanations of the marked differ- ence in the homicide rates of Negroes and whites. An excep- tion to this is Hooton ( 34.); who reported anthropological findings purporting to show, among other things, that crimi- nal homicide among the Negroes could be attributed to inher- ent racial traits. Harlan ( 29 ), in a study of 500 homi- cides committed in and around Birmingham, Alabama, emphati- cally contests the stand of Lombroso and Hooton stating that “rather the explanation is to be found in the life circum— stances of the Negro - the community setting in which he lives and the cultural tradition characteristic of his group life.“ Gillin ( 24 ) ( 85 ), in his study of the prison popula- tion of Wisconsin, which included 93 persons serving life terms for murder at the time, concluded that the Wisconsin murderer is a “Criminal by passion.“ He lists the basis for this conclusion as: only 2 out of 5 killed in relation to the perpetration of another crime, whereas 1 of 5 killed as the result of an immediate quarrel and 2 of 5 killed as a result of a long-standing grudge or marital difficulty. He also re- ported that the murderers were of generally lower intelli- gence than the other prisoners (he found 2 out of 3 to have I.Q's less than 70), were somewhat older, and tended to be more characterized by “abnormal mental conditions“'than other inmates. The basis of the latter finding - that of 1/4 of the murderers were either insane when committed or became in- sane after incarceration - would bear questioning, as it does not seem to agree very closely with the findings of other in- vestigators. In an intensive study of the case histories of these murderers, Gillin (34),noted that common to them was “(1) an incapacity to adjust normally to life situations; (2) harmful emotional experience in early life; and (3) a crisis too severe for an unstable person to cope with sanely.“ Hodgdon ( 33 ), studied the social characteristics of 400 first degree murderers in Michigan prisons and came to 10 the usual conclusion that, as he puts it, “there is no pana— cea for crime.“‘ He found for the great majority of murderers he surveyed that in each case many factors - instead of any single factor - was involved. The main factors he found in- cluded the facts that: murderers as a group were considerably older than members of the general prison population; “alco- holism“ appears to be associated with the crime; an unduly high proportion of single men, and divorced and separated men were included in this group (Hodgdon interprets this as in- dicative of an over-all maladjustment on the part of the in- dividual concerned); the broken home seemed to be an impor- tant factor in the cases of whites, but not for Negroes; the educational status of these murderers were found to be low; and 20.4% of them were found to have intelligence quotients in the feeble minded range (below 69) as compared to 3.1% for the general male population (as reported by draft statis- tics.). He lists the following motives as present in the 400 murders he studied: profit or personal gain, 141 cases (35.3%); non-family altercation, 90 cases (33.5%); family altercation, 79 cases (19.8%); other motives, 53 cases (13.3%); result of another crime, 30 cases (7.3%) and re- sisting arrest, 13 cases (3.0%). Frankel (31), in his study of 1000 murderers in New Jer- sey concluded that society must learn to control the human passions in order “to prevent the needless slaughter of such large numbers of fellow human beings“. However he, like many 11 other investigators of the causes of homicide, feels that “deeds are not the result of rational processes, but are de- termined instead by all kinds of jumbled instincts, impulses and desires - conscious and unconscious - breaking through the inhibitions imposed by the individual upon himself or the external pressures exerted by the group of which ha’forms a part“. He concludes that before control can be achieved by man, a great deal of scientific work needs to be done which will enable us to understand the murderer and how he comes to be as he is. It is interesting to note the mental diagnoses offered by psychiatrists who examined 731 of the 1000 murder- ers that Frankel studied: Table II MENTAL DIAGNOSES (PSYCHIATRIC) OF 731 MURDERERS —__———¢- -—‘-“- —— V_ ..H ' ‘__- _‘_-A Diagnosis Number Per Cent All individuals 731 100.00 No psychiatriccondition 383 53.1 Constitutional defective 109 15.1 Chronic Alcoholic 135 18.7 Epileptic 5 0.7 Psychopathic 41 5.7 Psychotic 48 6.7 13 3. Psychologically-oriented studies Cassity ( 13 ), in his “personality study of 300 murder- ers“ concludes that frustration in sexual, social, or econo- mic striving is unquestioningly the "prime motivating ele- ment of the homicidal act.“ He feels that the economically and socially frustrated are avenging themselves on the world for their sorry plight and that their hostility represents a protest against deprivation directed originally against the parents, but later enlarged upon to include all authority. As an example of this type of murderer he cites those who kill during the course of robbery or larceny, most of whom he says are of borderline or dull normal intelligence and have dis- played an inability to compete since childhood. He includes under the heading of “due to frustration in sexual strivings" the cuckold murderers, murderers of parent surrogates, murder in the course of homosexual panic and certain alcoholic mur- derers. Of this entire group of 300 murderers, only 3% were found to be feeble minded and only 3&% were found to be in- sane, a proportion somewhat smaller than that reported by other investigators of murderers, but which he reports as be- ing in line with Cromberg and Thompsonos general survey of 10,000 clinic cases. Berg and Fox ( 7 ), in a study of 300 male first and second degree murderers at Southern Michigan State Prison suggested that the lower intellectual adequacy that they found for the homicide group (as compared to the general pri- 13 son population) may eXplain how those murders which culmi- nated from an argument may have come about. They felt that in such cases as these the argument was characterized by a degradation from a verbal to an emotional, and finally to a physical level at which time the murder occurred. In other cases, such as those who murdered women, they felt that an adjustment in a problem situation was required of the indi- vidual which he could not make, perhaps because of the low- ered intelligence and lack of education as well as other factors, with an altered scale of personal values being in» volved in such instances. A more direct analysis by these authors of the various factors related to homicides also in- dicated that men who slew females were significantly older than those who slew males; and that when females were slain, the slaying tended to be done in a violent manner and “ago“ was often involved in the slaying. The general trend noted for this group was for younger men to murder for economic gain or to avenge an insult and that they tended to kill older men. The relatively older men tended to slay women for motives related to infidelity or arguments which were really over money. The tendency to consider most murderers as psychopathic or psychotic doesn't appear to be borne out to any significant extent in the data encountered by this writer. Henderson ( 33 ), reports finding only four psychopathic murderers out of the 49 he has examined. He describes all of these four as 14 being egocentric, individualistic types who had never shown any consideration for others and who had always regarded themselves as a law unto themselves. It would seem that it is individuals such as these egocentric, asocial “psychopaths" and the cold, apathetic, schizoid individuals who are so de- void of human empathy (such as Gilbert ( 36 ), describes S. S. Colonel Roses and other Nazi concentration camp executives as being) that prompted the early criminologists to speak of the “moral degeneration of murderers to a sub-human level.“ The occasional senaationally publicized occurrence of some individual apparently going psychotic and “running amok" with a gun or a knife has led the public to believe that the insane murderer is an exceptionally dangerous individual and that every psychotic person is a very dangerous potential murderer. Cruvant and laldrop ( 16 ), in reporting on 81 murderers admitted to St. Elizabeth's Hospital stated that homicide is really a rather infrequent consequence of mental disorder and that mentally ill people who commit a homicide do not constitute a psychiatric entity as distinct from other mentally disordered patients, but respond in the same way and in the same degree to identical methods of care found useful to mentally ill patients in general. Kurland, Morganstarn and Sheets ( 41 ), in a study of a selected sample of 13 wife- murderers who were subsequently admitted to a state psychi- atric hospital, emphasized paranoia as a prominent feature of this group. They suggested in this regard that the conflicts 15 produced by latent homosexuality might have played a large part in the behavior of these individuals. 11. PROBLEM It is apparent that the studies of murderers that have been published to this date either tend to be very highly in- dividualistic or widely generalized in nature. Their authors are essentially in agreement that murderers are maladjusted individuals but make little attempt, other than occasionally on an individual level, to ascertain the aspects of the mur- derer's personality that makes for such a serious maladjust- ment. Most of these authors have been content to point out the generally inadequate socio-economic and hereditary back- ground of the imprisoned murderer, along with the finding that the average murdererms level of intelligence tends to be lower than that of the average of other imprisoned individ- uals. Several ( 3, 13, 34, 35 ) have noted characteristics of murderers which would indicate that such traits as imma- turity and low frustration tolerance are common to them as a group, but the only psychological test that has apparently been used to any extent has been the intelligence scale. It is sur- prising, considering the psychologist's interest in research concerning human behavior and the great number of tools that he has devised to evalute the human personality, that no psychologist has undertaken the task of applying some of these tools to a study of these individuals who have taken it upon themselves to kill another. Despite Banayos ( 3 ), conten- tion that anygng may be a murderer under proper circumstances 16 this author is inclined to agree more with Karpman's state- ment that: There must be differences, specific differences indeed, that explain why, in a seemingly similar setting, one individual will commit a murder and the other will not, (but) will leave his wife and get a divorce. Another will commit murder and follow it by suicide; and still another will merely commit suicide but will not attempt mur- der at all. But there are many other types, the psychodynamics of which are virtually unknown to us. ( 36, p.40) Karpman, who could be considered America's foremost “criminal psychiatrist“, has published a considerable a- mount of data on murderers whom he has treated by psycho- analysis. As a result of his work in this area Karpman ( 35 ), notes, along with Gillin ( 35 ) that practically all murders may be loosely termed “crimes of passion“ in that the motives behind them are emotional ones. Karpman exempts from this loose passion murder category only what he calls “the strictly predatory murder“, that is, the murder which is premeditated in the hope of gaining the victim's money or property, and the murder incidence to a holdup or robbery. He feels that even in this “strictly predatory murder“ cate- gory there are several types: One group was not unlike the common hard—boiled criminal, whose analysis fails to reveal an o- vert emotional background. Yet, on careful 1? analysis, such is often found. In another type of murder, behind the apparent motives of gain one finds inner and more deep-seated motives of which murder is only a superficial, sometimes even an accidental expression. Still another group is the purely psychopathic.... which in- volves a minimum of emotional content, a callous and appalling disregard for human life and apparently no evident emotional aspects. It is as it were, a total lack of normal generous hue man emotions. (36, pu40 ) It is difficult to disagree with Karpman,s assertion that intensive (1.6., psychoanalytic) study of the individ- ‘ual murderer is the most certain way of uncovering his pay- <3hodynamics but it should not be unreasonable to formulate the hypothesis that there may be some common factors in the personality structures of individuals who commit murder, especially so in the case of those who commit murders with a 1milar motivation (e.g., armed robbery, suspected infidelity, etc.). With this statement, and the implicit assumption ‘tllsrt those who commit murder are quite probably not normal, We arrive at the focus of this study: the utilization of DOYchological tools to ascertain the personality structures 0f two groups of convicted and imprisoned murderers. The author's original plan was to do a comparative study between a group of convicted and imprisoned murderers 18 and a number of inmates of the same prison convicted of lesser offenses such as embezzlement, forgery and such crimes against property. However, the finding in the literature of the con- jecture that there are two main types of murder, so-called “crimes of passion“ and “crimes for profit“, led the author to look closely at the available subjects -- the inmates of the prison system of the State of Michigan. It was found to be readily possible to divide the great majority of those convicted of first and second degree murder into two groups; one consisting of men who had actually murdered their wives and the other of men who had actually murdered in the course of robbery. It was from these two groups that the author drew the subjects for this study. For the purpose of an ex- ploratory study, it was decided to use one objective and one ‘projective test to provide clues to personality differences and similarities between these two groups and between them and the control groups. 19 III. METHODOLOGY A. Sample The population used in this study consisted of a total of 40 convicted and imprisoned male murderers, 35 of whom were convicted of murder in the first degree, and 5 of whom were convicted of, or plead guilty to, murder in the second degree.1 All of these men were confined in the Michigan State Prison System; 33 at Southern Michigan State Prison at Jackson, Michigan, and 7 at the Branch Prison at Mar- quette, Michigan. To minhmize the effect of “institution- alization“, the group as a whole was limited to men who had served less than 5% years on their present sentence at the time of testing for this study. To avoid the complications of mental retardation those who had not achieved an 1.0. of at least 80 on the Army “Alpha“ Test were eliminated. This test is routinely administered to each inmate by one of the prison psychologists shortly after he is received in the prison system. lin the State of Michigan, murder in the first degree is generally defined as “premeditated“’and intended murder, or murder committed during the commission of another crime (such as armed robbery or rape). “Murder in the second de- gree" implies lack of premeditation, but presence of intent such as the case of murder during a fight with another in— dividual. Most of the wife-murderers in this sample were given the Opportunity to plead guilty to second degree or even man-slaughter, but elected to fight their cases and were subsequently convicted of first degree murder in a jury trial. 30 Twenty of this group of 40 men were convicted of mur- dering their wives and thus would fit into the “passion mura derer“ category. The five man convicted of “murder in the second degree“ were in this group. Ten of this group were white men and ten were colored (Negro).men. This group, for the lack of a more convenient designation, will be here- after referred to as the “passion murder“ group. The other twenty were convicted of murdering during the course of robbery and thus could be placed in the “profit murderer“ category. These men were all convicted of first degree murder. Ten of this group were white men and ten were Negro men. This group will hereafter be referred to as the “profit murder“ group. It should be noted at this time that only men who were convicted of being the actual agent of the act of murder were included in the groups whose test data were used in this study. These groups were equated roughly on the basis of socio- economic status and time served in prison on their present offense. They were equated more exactly on the basis of age and intelligence (Army Alpha I.Q.'s), as can be seen in Table III. The “profit murderers“ had a much higher inci- dence of previous difficulty with the law, averaging almost two previous convictions each as compared to the “passion murderers“ average of less than 0.5 previous convictions per man. There was little difference in preference of weep pons with which the crime was committed, 19 of the “profit 31 murders“ and 18 of the “passion murders“ were committed by use of handguns, one in each group by use of knife and one man in the “passion group“'ran his wife down with the fami— ly car. In general, these men were very cooperative and easy to work with. Of the 48 men who were asked to take part in this study, only one refused. This man was a young Negro wife- murderer who refused on the grounds that he was doing all right at present and that there was nothing wrong with his mind. The cooperation of the other men ranged from quiet compliance to apparently eager interest. Much of this cooperation from these men -- who stood to gain nothing by taking these tests except possibly chiding from fellow inmates as being “bugs“ -- was quite apparently due to a desire to accomodate someone who seemed to be taking an interest in them as individuals.1 The reader has quite probably noted that the design of this study does not include a group of “normals“ or “con- trols“ matched with these two groups as to age, socio— economic background, and I.Q.. The most important, and most obvious, reason that this is true is the extreme diffi— 1G. M. Gilbert (private communication) also noted this in his work with the leaders of the 1953 Southern Michigan State Prison riot, and, in a similar manner, the Nazi war prisoners at the Nuremburg trials. 23 culty in locating such a group. A group of prisoners could have been obtained from the same prison system (and M.M.P.I. data from such a group of prisoners was one of the sets of data used as comparative purposes) but such men can hardly be considered as a yardstick of normality, and besides the avowed purpose of this study is to attempt to study the per- sonality structure of two groups of murderers by the use of thepsychological tools of the Rorschach and the M.M.P.I., not just to compare them with other convicted men. The author does not feel that due consideration of the data ob- tained from such a group would be within the scope of this study. Besides the established “norms“ (30) of the M.M.P.I. we have the above mentioned prison inmate data and the data from a group of “normal“ subjects published by Schmidt ( l4 ), available as controls. 0n the Rorschach we have the data published by Rapaport ( 48 ), on his “well-adjusted“ Kansas State patrolman and Lindner's “normal“ prisoners ( 43 ). B. Test Matgrials and ngcedurg Two psychological tests, the Rorschach Technique and the Minnesota Multiphasic Personality Inventor, were given by the author to each of the 40 individuals used in this study. The Rorschachs were administered individually. Both administra- tion and scoring was done according to Back ( 4 ), but Klopfer's animal movement ( 39 ) was scored in addition. The Paper and pencil form of the M.M.P.I. was used. A third test, The Michigan Sentence Completion Test, was given to the first eighteen men examined, but discontinued because inspection of the results obtained by its use revealed that it was too strongly influenced in this case by the individual‘s present situation to be useful in this study. The Rorschach and the M.M.P.I. were selected for use in this study for several reasons: for one, they are quite widely used in psychological practice and there is a modicum of agreement as to the meane ing of the material obtained by their use. Also, they are amenable to quantification and thus statistical evaluation of differences between populations are possible. Other ad- vantages include the facts that they have been quite exten- sively used in previous research, with the result that there are normative groups available for comparative purposes; that the M.M.P.I. is objectively scored, and that the Rorschach scoring is such that subjective errors should be minimal and scoring reliability high with a well-trained scorer. For the administration of these tests, the subjects were seen in a private office within the walls of the prison in which they were presently incarcerated. When additional office space was available, the subjects were seen in groups of two or three, with one or two of them working on the M.M.P.I. while one was being given the Rorschach in another office. Cigarettes were made available at these times to aid the subjects in feeling more at ease in the situation. The subjects were never officially told the purpose for their be~ ing called to this office until they met the author, although it is very probable that the majority of them had heard a ru- 34 nor via the prison “grapevine“ as to what was taking place. Upon being seen in the private office by the author the sub- jects were told that their cooperation was desired in a study being done by the author in an effort to find out just what sort of individuals were imprisoned for murder. They were given to understand that they were not required to take the tests being used in this study if they did not want to, that none of the prison officials would care whether or not they took them, and that all individual findings would be kept strictily confidential to the author and his thesis advisor. A-more exact and detailed set of general instruc- tions will be found by the reader in Appendice #1. As pre— viously noted, the subjects on the whole responded well to the situation with only one refusing to take the tests, despite the fact that they were all informed before-hand that they quite probably stood to gain nothing themselves from taking the tests. 35 IV. RESULTS A. Rorschach Findings Table III contains background information as to the age and obtained Army Alpha I.Q.'s of the two groups of convicted murderers, along with information as to the age of the twenty youngest of Rapaportos ( 48 ), “Nell-Adjusted“ Patrolman group. There were thirty-four men in this group of Rapa- port's but this author used the Rorschach data of only the twenty youngest (age range 35 to 37, inclusive) in order to have an identical N with the two murderer groups and to equate the three groups as closely as possible as to age. 1 A summary of the Rorschach scores on each of these groups is I listed in Table IV, along with the Rorschach findings on Lindner's ( 43 ), “normal prisoner“ group. A nonpparametric method, the Kolmogorov-Smirnov test ( 37 ), was employed in analyzing the data of the two murder- er groups and the Rapaport group. This test utilizes the in- dividual scores within each group and since individual scores were not available on Lindner's group it could not be included in this statistical analysis. This Kolmogorov-Smirnov Test is relatively new and is considered to be more adequate than the older, chi-square method which has usually been utilized in evaluating data that are not continuous and not normally distributed. The two-tailed form of this test was used, as the results were not predicted, and only differences at or greater than the probability level of .034 ( 43 ) were re- 36 ported as significant. Table V presents the results of the comparisons of: (l) the “Passion“ murderer group with Rapaport's group; (3) the “Profit“ murderer group with Rapaport's groups (3) the “Pas- sion“ group with the “Profit“ group; and (4) the “combined“ (both the “Passion“ and “Profit“'group) murderer group with Rapaportus group. There were no significant differences found between the two murderer groups“ Rorschach scores, but the incidence of FM, CF, and 8 were found to be significantly greater for the “Passion“ group than for Rapaport's group, and the FM incidence to be significantly greater and the FC incidence to be significantly smaller for the “Profit“ group than for Rapaport's group. lhen the Combined murderer group was compared with Rapaport's group, FM, CF, and S was found to be greater and FC smaller for the Combined murderer group. It should also be noted that there is a consistent, though not significant by our standards, trend for M and l to be less and D to be greater for the murderer groups than Rapaport's' group. The only difference between the two murderer groups that approaches significance is that of the “Passion“ group showing more CF than the “Profit“ group. B. Minneggta Multiphasig Personaligy Tgst Finding! Tables VI through XI inclusive, are concerned with the M.M.P.I. findings, which are presented both in raw score and scaled score form. The statistic used here to analyze the 37 data is Fisher's “t“ for differences between means as given by Guilford ( 38 ). The formula used allowed for combining the sum of squares of the group scores being tested and the use of the formula (N1 { N3 - 3) for ascertaining the number of degrees of freedom with which to enter the T-table. The exception to this is when the variances of the two groups differ significantly according to the F-Test for variance ( 38 ) in which case this author followed Cochran and Cox's ( 15 ), suggestion to use the formula N1 - l to determine the number of degrees of freedom, with N1 being the number of subjects in the smaller of the two groups being compared. It is becoming more the practice when subjecting the M.M.I.I. to statistical test to use the raw scores rather the scaled scores ( 31 ). This author was unable to obtain the individual raw score distributions or even the raw score means or standard deviations for the only non-prison inmate group that he had available ( Schmidt's 98 “Normal“ soldiers) so two sets of statistics were run, one on the scaled scores of all of the groups and the other on the raw scores of all of the groups except Schmidt's “Normals“. Actually, as will be seen, there was very little difference in the findings of the two methods. Table VI presents the M.M.P.I. scaled score means and standard deviations for the murderer, general prisoner and 38 Schmidt groups. The M.M.P.I. categories1 are listed on the left boundary of the Table with the “Passion“ murderer group scores falling in Column 1, the “Profit“ murderer group scores in Column 3, General Prisoner group scores in Column 3, Schmidt“s “Normals“ group scores in Column 4 and the Combined murderer group scores in Column 5. A graphical illustration of thee scores is given as Table VIIa and VIIb which are standard M.M.P.I. “Profile Charts“ on.which the mean scores of the General Prisoner, the two Murderer and Schmidt's Normals groups are located for ease in visual comparison. That there is somewhat more similarity than difference between the performance of the two murderer groups on the M.M.P.I. is shown by the comparisons between these groups and the two control groups, both individually (Table VII) and as a “Combined Murderer“ group (Table VIII). The significant differences between the two murderer groups were limited to the areas of K, F, and Si, with K showing a p at the .05 level of being greater in the “Passion“ group, and F with p at the .01 level and 81 at the .05 level being greater in the “Profit“ group. There was enough of a mean difference in (dubious) favor of the “Profit“ group in the areas of Pd and Mf to re- sult in t's that approached the 5% level of significance and possibly minor trends in the areas of D, So and Ma in the 1For an interpretation of the signs used to designate these and a short discussion of their meanings please see Appendice 3. 39 same directions as the “Profit“ group showed a somewhat generally elevated profile over the “Passion“ group. The “Passion“ group, in turn, has generally higher mean scores than the General Prisoner group, but here there are only two scores, those of Hy at the .01 level and K at the .05 level, that prove to be significantly different in comparing the two groups. The areas of He and D show a fairly strong trend toward significance in the direction of the “Passion“ group over the General Prisoner group, but the rest of the scores show generally less differences than the above-mentioned “Passion“ vs. “Profit“ group comparisons. Comparison of the “Profit“ and the General Prisoner groups' mean scores (Column 4, Table VII) show the “Profit“ group mean scores to be significantly larger for F, 81, D and Hp; The differences between the two groups on F and Si were significant at the .01 level; D and Hy at the .05 level. Although Mf is the only other category that approaches signi- ficance, it should be noted that all of the categories other than K, L and Pa show a mean difference of 3.3 or greater on the side of the “Profit“ over the General Prisoner group. When the “Profit“ and “Passion“ groups together, as a “Combined Murderer“ group, are compared with the General Prisoner group (Table VIII, Column 1), their common differ- ences with this group in the categories of Hy and D are accentuated so that they both reach the .01 significance level. The other significant differences between the “Com- bined.Murderer“ group and the General Prisoner group are F and Si (at the .05 level) both of which receive most of their weight from the “Profit“ group component of the “Com- bined“ group. There were no other mean differences between the “Combined Murderer“ and the General Prisoner group that can be considered as approaching significance, but it should be noted that these mean scaled scores for all of the M.M.- P.I. categories were larger in the case of the “Combined Murderer“ group than the General Prisoner group. Comparison of the Murderer groups with Schmidt's “Nor- mals“ (Table VII, Columns 3 and 5; Table VIII, Column 3) result in an almost categorical finding of the Murderer group having significantly larger (.01 level) mean scores. The only exceptions to this (other than the K and Si scores, which were not reported by Schmidt) were: (1) in the “Pas- sion“ - Schmidt comparison (Table VII, Column 3), where F was not significant and L was significant at only the .05 level; (3) in the “Profit“ - Schmidt comparison (Table VII, Column 5), where L was not significant; and (3) in the “Combined Murderer“ - Schmidt comparison (Table VIII, Col- umn.3), where L was not significant. The significant find- ing in the case of L in the “Passion“ - Schmidt comparison comes in the face of only two scaled score points difference between the means of these groups at a point where they are both close to the standardized test mean of 50. As can be seen in Table VI (Column 5) many of the mean scaled scores of Schmidt“s group fall below the M.M.P.I. “standardized 30 31 mean“ of so, quite in contrast to the “Combined Murderer“ group, whose mean scaled scores range from a low of 53.35 (for Si) to a high of 71.3 (for Pd). In the M.M.P.I. raw score comparisons of the murderer groups and the General Prisoner group (Tables X and II) the findings were almost identical to those obtained by use of scaled scores. In the “Profit“ - General Prisoner compari- sons D came out a little below the .05 level of significance and as a result also slipped to the .05 level in the “Com- bined.Murderer“ - General Prisoner comparison. The T for mean differences in the “Combined Murderer“ - General Pris- oner comparison on He was somewhat greater here than with the scaled score comparison, but not enough so to reach significance. In general, it should be noted that both the “Passion“ and the “Profit“ murderer deviate significantly on all of the clinical scales of the M.M.P.I. when compared with a “Normal“ group and on three of these scales (Hy, D and 81) when com- pared.to a general prisoner group. As will be shown in the following discussion, not all of the significant findings be- tween.the “Normal“ and Murderer groups can necessarily be con— sidered as diagnostic of deviant behavior, but a good many points can be made in this direction on the basis of these M. M. P. I. findings. Table III AGE AND I.Q. OF TWO GROUPS OF CONVICTED MURDERERS. BO‘YOUNGEST 0F RAPAPORT'S “WELL-ADJUSTED PATROLMEN“. 33 ACE OF Group N Age 1.0. M SD M SD “Profit“ Murderers 30 30.48 7.65 91.85 8.48 “Passion“ Murders 30 33.71 8.05 94.60 9.51 Patrolman“ Table IV 33 MEAN SCORES ON THE RORSCHACH TEST FOR THE TWO MURDERER GROUPS, RAPAPORT'S “WELL-ADJUSTED“ PATROLMEN GROUP AND LINDNER'S “NORMAL“ PRISONER GROUP (1) (2) (3) (4) (5) “Profit“ “Passion“ Rapaport “Well Lindner Combined Murder Murder Adjusted“ Patrol- Prisoner Murder men “Normals“ (N = 20) (N = 20) (N = 20) (N 2 40) (N = 40) M S.D. I S.D. M S.D. M S.D. M S.D. R 18.35 7.66 18.85 6.44 18.95 8.90 17.8 11.7 18.55 6.71 W 5. 30 4. 37 6. 10 3. 30 7. 75 3.93 5. 7 3. 5 5. 70 4. 16 D 12.0 5.10 12.05 4.89 9.50(1) 6.39 10.6(1) - 12.02 4.95 .mi 0.85 1.31 0.70 0.96 1.70 3.04 1.4 1.8 0.76 1.33 .M 0.75 0.73 0.90 0.79 1.35 1.37 1.5 1.4 0.83 0374 FM. 3.00 3.45 1.40 0.80 0.30. 0.55 4.61 4.3‘ 1.70 3.01 FC 0.30 0.41 0.35 0.49 1.30 1.34 0.9 1.3 0.38 0.46* CF 1.65 1.73 3.40 1.31 0.85 0.81 0.6 0.8 3.03 1.38 C 0.35 0.49 0.15 0.11 0.05 0.05 0.5 0.3 0.35 0.32 Y 1.58 1.94 1.38 1.31 1.85“ 1.77 1.8“ 1.3 1.48 1.86 H 0.85 0.95 1.35 0.91 3.35 3.58 1.5 1.3 1.05 0.94. (H # Hd)(H l Hd 0.70 0.79 0.60 1.37 . 1.4 1.9 0.65 1.14- 5% 3.60 9.16 45.90 6.44 51.7 5.43 47.2 - 49.25 8.03 An 1.30 1.87 0.75 1.17 1.30 1.15 1.0 1.3 1.03 1.71 s 1.95 3.01 1.30 1.13 0.35 .83 0.6 1.3 1.63 1.33 P 4.50 1.39 5.50 1.61 N.C. - N.C. N.C. 5.00 1.47 F% 2.9 31.1 67.7 27.0 - - 35.0 33.6 70.5 38.9 FM 2.5 13.6 80.5 12.9 — - N.C. N.C. 76.5 13.4 (1) Klopfer's “D“, plus his “d“ N.S. Not scored or not reported “ Klopfer's Fc N.C. Not comparable, due to wide differences here between Beck's and K1Opfer“s scoring systems. 34 Table V MAXIMUM DIFFERENCES - PROBABILITY VALUES OBTAINED BY USE OF THE KOIMOGOROV-SMIRNOV METHOD OF INTERGROUP COMPARISON ON THE TWO MURDERER GROUPS AND RAPAPORT'S WELL ADJUSTED PATROLMEN GROUP (W. A. P.) (IT (3? TU (47 Rorschach “Passion“ “Profit“ “Profit" Combined Murder Score vs. W.A.P. vs. l.A.P. vs. “Passion“ vs. W.A.P. d d d d _‘ W -.35 -.40 .35 .375 D .30 .40 .35 -.335 Dd -.15 -.35 -.10 -.l75 u -.30 -.30 -.10 -.300 ;1 FM .65“. .75“ .15 -,475** F0 -.35 -.50*‘ -.15 -.435* CF 0 60.. c 25 -0 35 o 435. C .10 .30 .30 .350 EY -.30 -.35 .30 -.375 H { Rd -.35 -.30 -.10 -4175 A-# Ad -.15 -.30 -.30 -.335 An. -.35 -.30 -.15 -.335 3 .45‘ .40 .35 .435‘ P N.C. N.C. -025 N.C. N.C. Not comparable due to differences between Beck and Klopfer scoring systems. ‘ Significant at or beyond the .034 level of confidence. “ Significant at or beyond the .013 level of confidence. Table VI 35 MEAN SCORES ON THE M.M.P.I. FOR TWO MURDERER GROUPS} BOTH INDIVIDUALLY AND COMBINED, A“GLNERAL PRISON“ GROUP, AND SCHMIDT'S “NORMAL“ GROUP. w-— (1) fl—m'v—v— . “-vv 9 (4) “Passion“ “Pré§2t“ Geéggal Schmidt's Coéggned Murderers Murderers Prisoners “Normals“‘ Murder N n-30 N -’30 N = 30 N = 98 N = 40 Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean S.D. K 59.3 9.5 53.0 8.9 53.8 8.9 - - 56.15 9.7 F 53.3 4.7 67.0 13.3 55.0 5.4 53.3 4.1 60.1 8.8 He 56.3 9.6 56.4 10.1 53.3 9.3 48.5 8.3 56.3 10.3 D 59.0 6.3 63.0 11.4 54.3 9.5 41.3 13.7 61.0 8.8 Hy 61.4 8.7 59.8 7.8 53.5 8.8 45.1 10.0 60.6 9.1 Pd 68.4 10.9 74.2 9.6 71.0 8.1 47.0 13.1 71.3 10.7 If 53.8 7.3 59.3 10.1 54.0 7.8 45.3‘ 9.7 56.55 10.5 Pa 57.0 10.0 59.4 7.7 57.1 8.3 51.0 8.9 58.3 10.1 Pt 56.0 8.9 54.3 9.6 54.3 10.1 49.5 8.4 55.1 9.7 8c 58.4 8.7 63.7 16.3 57.3 10.6 48.5 7.7 60.55 13.0 Ma 57.7 10.3 63.8 14.5 58.3 13.4 48.3 9.4 60.35 13.6 81 49.5 7.3 55.0 9.0 46.8 7.15 - - 53.35 9.3 L 54.1 4.8 53.3 3.9 53.8 4.3 53.1 3.5 53.3 4.5 36 Table VIIa THE P.I. PROFILES FOR THE “COMBINED“ MURDERER GROUP, ILL(AENERAL PRISONER GROUP AND SCHMIDT'S “NORMAL“ GROUP. PROFILE CHART l 10 100 90 80 '1' Score 60 50 40 30 20 10 21.x? H.0H,p.M.p.p.s.M. +3: +A! +1: +1: +3! , Standard 4 ('1')Sccre____ _ __ _ .... __ _ _ __ _ _ ( RawSccre_._.—_ _______._.___._ Ktcbeadded_ __ _ _ __ RchccrewithK_ _ __ _ _ r I. ._____. Combo'xcfl vahrlv 8"”A‘ (N’4°) 1. '---—~ 9fieneVs‘ rw'sswu ”“8“"? (”‘1‘0 J. -—~~-—- xekmn's'reumus” (N214) 37 Table VIIb M.M.P.F. PROFILES FOR THE “PROFIT“ AND THE “PASSION“ MURDERER GROUPS. PROFILE CHART 12071.1(? H.DH,P¢M.P.P.S.M.5& 110 100 71.x? 11.1)11.15.149.143.14.5.7~ +.5x +.4x +1: +1: +3: Standard ('1') Score _ _ _ __ BawSccre____ ___________ Ktcbeadded... _ _ _ __ RawSccrewithK_ _. _ _ _ A. ...—..onPHYITaAMWAlv‘V 3”“? (than) 2- """"‘" ..PaSSiuV\"Mvwd ¢T¢V bV'Kf (“’19) .--_-1-R" “ r ' 38 Table VIII MEAN DIFFERENCES AND T-VALUES FOR DIFFERENCES ON 11.11.13. I. SCALES BETWEEN THE TWO MURDERER GROUPS, A GENERAL PRISONER GROUP AND SCHMIDT‘S NORMATIVE GROUP. “Pass ion“ Murd. “Pass ion“Murd. “Passion “Murd. “Profit “Murd. “Prof it“Murd. with with with 1with with “Profit“Murd. General Pris. “Normals“ General Pris. “Normals“ I diff t M diff t M diff t M diff t M diff t K 6. 3 3. 188* 6. 5 3. 333“ - - 0. 3 0.071 - - F -13.8 4. 357” -1.8 1.131 1.0 0.971' 13.0 3.767" 14.8 6.554 Hs -0.3 0.064 3.9 1.976 7.7 3.735” 3.1 1.014 7.9 3.765 D -4.0 1.380 4.8 1.884 17.8 5.993" 8.8 3.653“ 31.8 7.113" Hy 1.6 0.617 7.9 3.854" 16.3 7.111" 6.3 3.405 14.7 6.303" Pd -5.8 1.784 -1.6 0.538 31.4 6.833" 3.3 1.141 37.3 8.835M Mf -5.5 1.974 ~0-3 0.083 8.6 3.504“ 5.3 1.854 14.1 5.887" Pa -3. 4 0.847 -0. 1 0. 041 6'. 0 3. 688" 3. 3 0.904 8. 4 3. 935" Pt 1.8 0.613 1.8 0.597 6.5 3.130" 0.0 - 4.7 3.333" 80 -4.3 1.045 1.3 0.391 9.9 5.518" 5.5 1.373 14.3 6.016” Me. -5.1 1.389 -0.5 0.139 9.5 4.055" 4.6 1.084 14.6 5.733" Si -5.5 3.077“ 3.7 1.194 - - 8.3 3.191" -— - L 1.8 1.300 3.3 1.773 3.0 3.288“ -0.5 1.096 0.3 0.333 " Significant at the .05 level of confidence. ** Significant at the .01 level of confidence. Table IX 1 DIFFERENCES AND T-VALUES FOR DIFFERENCES ON M.M.P.I. SCALES NEEE THE COMBINED MURDERER GROUPS, THE GENERAL PRISONER GROUP AND SCHMIDT'S NORMATIVE GROUP. (1) (2) Combined Murderers. Combined Murderers with with General Prisoners Schmidt "s “Normals“ M diff t M diff t Y. 3. 35 l. 389 - - F 5.10 3.373‘ 7.9 4.633" '83 3.90 1.063 7. 7 9.000" D 6.80 3.739" 19.8 8. 479“‘ By 7.10 3.891" 15.5 6.114" Pd 0.30 0.110 34.3 13.434" Rf 3.55 0.958 14.35 4. 143" Pa 1.10 0. 433 7.3 3.393” Pt 0.90 0.334 5.6 6.735" 80 3.35 0.996 13.05 6.145" la 3. 05 0. 594 13 . 05 7. 013" Si 5.45 3. 305“ - - L 0.40 0.339 1.1 1.534 3 Significant at the .05 1eve1 of confidence. #3 Significant at the .01 level of confidence. 40 Table X M.M.P.I. RAW SCORE MEANS AND S.D. '8 FOR THE TWO MURDERER GROUPS, BOTH AS SEPARATE GROUPS AND AS A COMBINED GROUP, AND THE GENERAL PRISONER GROUP “Paéizon“ “Prégit“‘ 6333331 Ccméiged Murderers Murderers Prisoners Murderers (N = 20) (N a 20) (N : 20) (N z- 40) Mean S.D. Mean 8 Mean Mean L 5.15 3.44 4.75 3.37 4.95 3.34 4.95 3.93 F 4.10 1.97 7.85 4.63 4.95 1.87 5.98 4.53 K 17.30 5.71 14.00 3.84 15.95 3.81 15.60 5.07 Hs‘ 13.75 4.38 13.95 4.43 11.73 4.30 13.85 4.06 D 30.35 4.11 33.15 5.91 18.10 5.13 31.35 3.96 Hy 33.50 5.17 31.85 4.88 18.15 5.31 33.18 5.13 Pd‘ 36.65 4.70 38.95 4.33 38.10 4.03 37.80 4.43 Mf 33.45 5.40 34.85 5.51 33.65 5.44 33.65 5.48 Pa 10.35 3.48 11.30 3.83 10.35 3.53 10.78 3.63 Pt‘ 35.70 4.44 35.30 5.43 34.90 6.06 35.45 4.89 So“ 36;55 4.63 38.60 . 8.71 36.50 5.06 37.58 6.96 Mt’ 19.65 4.56 33.05 5.73 30.10 4.61 30.85 5.08 Si 34.60 6.67 39.80 9.33 31.90 7.15 37.30 8.85 t BCOI‘OI o On these scales the K correction has been added to the raw ('oL’MIQ‘JJK‘ Ii- 41 Table XI MEAN DIFFERENCES, S.E. 0F MEAN DIFFERENCES AND T-VALUES FOR DIFFER- ENCES ON M.M.P.I. RAW SCORES BETWEEN THE TWO MURDERER GROUPS, BOTH AS SEPARATE GROUPS AND AS A COMBINED GROUP, AND A GENERAL PRISONER GROUP. (1) (3) (3) (4) “Passion“ “Passion“ “Profit“‘ Combined Kurd. with with with with “Profit“ General Pris. General Pris. General Pris. (N z 20) (N = 20) (N 20) (N a» 40) M diff “t“' M diff “t“ M diff “t“' M diff “t“‘ L 0. 40 0. 439 0. 30 0. 186 -0. 30 0. 318 0.00 - F -3s 75 as 333'. -0. 85 lo 403 as 90 as 598‘ 1.03 lo 000 K 3.30 3.308“ 1.35 0.864 -1.95 1.613 ~0.35 0.373 as -0.30 0.144 3.03 1.473 3.33 1.611 3.13 1.876 D‘ -1.80 1.338 3.35 1.497 3.05 1.747 3.15 3.647* Hy 0.65 0.409 4.35 3.6518 3.70 3.330”l 4.03 3.858" Pd -3.30 1.638 -1.55 1.119 0.85 0.653- -0.30 0.358 If -3.40 1.394 -0.30 0.117 3.30 1.373 1.00 0.667 Pa -0.85 0.735 0.00 - 0.85 0.733 0.43 0.436 Pt 0.50 0.319 0.80 0.476 0430 0.165 0.55 0.379 Sc -3.05 0.930 0.05 0.033 3.10 0.933 1.08 0.617 Ma -3}40 1.454 —0.45 0.315 1.95 1.189 0.75 0.564 81 -5.30* 3.036 3.70 1.335: 7.90 3.005“I 5.30 3.334“ it. Significant at the .05 level of confidence. Significant at the .01 level of confidence. 42 V. DISCUSSION A. Rorschach Test Findings The author does not intend to imply that he considers that a Rorschach summary of independent “signs“ such as is listed for each of the two murderer groups in Table III should be considered as a diagnostic entity of the groups which they represent. The intention here is to compare the Rorschach performances of the two murder groups with each other and with a control grbup of “normals“ (in this case, Rapaport's Well- Adjusted Patrolman), and to interpret these findings. The ”sign“’approach admittedly has many drawbacks (and extremely rare is the Rorschach user who interprets a Rorschach proto- col without content and sequence analysis) but it is not with- out value and does lend itself to statistical comparison. As a point of departure in the discussion of the Rorschach findings of these groups, it should be noted that the mean number of responses (R) for each of the two murderer groups, Rapaport's group, and Lindner"s prison group is closely iden- tical. Beck ( 4 )5 considers an R of about 35 to 35 to be “average“ and found his group of “normals“ ( 6 ) to have a mean R of 33.65. Thus, the obtained'R's of the four groups in Table IV, which range from 17.8 to 18.95, are quite be- low normal expection. Rapaport ( 48 ), explained the mean R of 18.95 for his Well-Adjusted Patrolman group as apparent- ly due to the “inhibited background“ of these subjects, im- plying by this that the “smalltown“ and rural background of 43 these men caused them to be inhibited and thus respond as such on the Rorschach. By the same token, the two murderer groups and Lindner's “normal prisoner“ group can be considered to be “inhibited“, but whether this was due mostly to their pre-pri- son back-ground or to their being prisoners and examined as such is open to question. In view of the many studies in re- cent years showing that the situation and the examiner can significantly affect productivity on the Rorschach ( 18, 38), this examiner does believe that the situation of being in pri- son could have had a significant effect on R. However, it is fortunate for the author that the productivity of these groups was so similar in terms of the mean number of responses per subject. Many of the Rorschach signs have been found to be related significantly to R, ( 30. 44 ), making statistical comparisons between groups with significantly different R's quite difficult, if not outright untenable. The most significant findings on the Rorschach Test were found in the “determinants“ of FM, FC and CF. In both the “Passion“ and “Profit“ murderer groups FM (animal movement) occumed with significantly greater frequency that in the Well- Adjusted Patrolman group. Along with this finding of high PM we note a consistent trend for the incidence of M (human move- ment) to be lower for the two murderer groups than the Patrol- men.group. FM and.u are usually interpreted as having quite different meanings, with the occurrence of FM being considered as a sign of emotional immaturity ( 39, 55 ), whereas M has gained fairly wide credence ( 10, 11, 13, 13, 14), as a mea— 'u _‘;:1 \ «l'IIIIA . 44 sure of empathy. The absolute incidence of FM (for “Profit“ group 3.0, “Passion“ group 1.4) is not extremely high but does overshadow the incidence of M (M = 0.75 for “Profit“ group, 0.90 for “Passion“ group) so strongly that we can only con- clude that both murderer groups' emotional development as measured by FM and M is more on the level of that of a pre~ adolescent child than that of an adult. The murderer groups' responses to color (Rorschach's FC, CF and C),add further emphasis to the emotional immaturity noted in relation to their production of FM and M and also be- speak as to their emotional responsivity and control. FC (form primary, color secondary) responses are regarded among those who use the Rorschach as indications of emotional con- trol, whereas CF’(color primary, form secondary) and C (pure color response) responses are regarded as indications of im- maturity, heightened emotional responsivity and poor emo- tional control. Clark ( 13 ), notes that “a greater number of FC responses (two or more) generally correlate with good adjustment“, Beck ( 5 ), states that PC “requires feeling in time with that of others, and in fact, indicates understand! ing of others through the medium of feelings“, and Klopfer and Adnsworth ( 38 ), state that “PC responses indicate a ready control over impact without loss of responsiveness . . this controlled responsiveness implies that the individual can.respond with both feeling and action appropriate to the emotional demands of the situation.“ FC responses are 45 notably lacking in the Rorschach records of the murderer groups with an average of only 0.30 PO responses per man in the “Profit“ group and 0.35 per man in the “Passion“ group. Statistical comparison of the production of PC by these two groups with that of Rapaport's “Well-Adjusted“ Patrolmen (ave. FC t-1.30), although falling a little short of signi— ficance in the case of the “Passion“ group above, showed the “Profit“ group and both groups as a “Combined Murderer“ group to have produced significantly less FC. We can quite safely say that the murderer groups are significantly lack- ing in the attributes measured by Rorschach's FC. 0n the other hand, both of the murderer groups' produc- tion of CF considerably exceeded that of either of the “con- trol“ groups. The average production of CF for each man in the “Passion“ group was 3.40, with “Profit“ group 1.65, in Rapaport's “Well-Adjusted“ Patrolmen group 0.85 and Lindner's “Normal“ Prisoners 0.60. In this case, our method of sta— tistical comparison disclosed a significant difference in the amount of CF produced by the “Passion“ group as compared to' the Patrolman group (P = .013) and in the “Combined Murderer“ group as compared to the Patrolmen group (P = 0034). The difference in production of CF between the “Passion“ and the “Profit“ murderer groups approached the significant level; more closely so than did the difference between the “Profit“ and.Patrolmen groups. This greater amount of CF that was produced by the “Passion“'group than was produced by the “Profit“ group provided the only close approach to a signi— 101.9... ('11:. ..1‘. . 1': 46 ficant finding between these two groups on the Rorschach Test variables by our method of statistical comparisons. This high CF production by these murderer groups so far oversha- dows their PC and pure 0 production that it can safely be considered as showing the tone of their emotional responsi- vity. Beck ( 5 ), states that the CF response is character- istic for a highly labile reactivity; that “the essential state is that of easy irritability.“ Bochner and Halpern ( 8 ), note that CF responses are linked to “impetuousness“ and “lack of social consciousness“, and Klopfer and Ains- worth ( 38 ), state that “if CF responses appear in a psycho- gram without signs of adequate control, the implication is ~— that there is an impulsive, uncontrolled acting out of emo- tional reactions.“ There were few pure C responses given by either the con- trol or the murderer groups, the highest incidence being a 0.35 average pure 0 response for the “Profit“ group. It should be noted however, that this pure 0 production of 0.35 per member of this “Profit“ group was higher than their average production of 0.30 PO per man. Pure 0 responses are considered as indicative of a potential for adjustment at a “relatively low genetic level“ (6 to 8 years mental age) and for “anti-social acting out“ by Phillips and Smith ( 47 ). .Beck ( 4 ), also notes that pure 0 is “an infantile response mode“ and that “the adult with pure 0 in his Rorschach record is likely to be given to ungovernable im- 47 pulses....' One of the standard Rorschach measures is the “experi- ence balance“ (Beck) or Erlebkis Types (Rorschach). Tech- nically, this is the ratio of the number of M to the sum of the color responses (sum 0), where F0 is weighted 0.5, CF as 1.0, and pure C as 1.5. Psychologically, its meaning has been found in many cases to be somewhat more difficult to evaluate than as the “balance of introversive against extra-tensive trends, of fantasy living against outward ex- pression of feelings“ ascribed to it be Beck ( 4 ), but this is nevertheless the basis for its usual clinical interpreta- ’_‘f..4.r.‘ ...--1;. .- tion. The “sxperience balance“ of both murderer groups is almost exactly in the ratio of 1 to 3 on the extratensive side (“Passion“ group M =-0.90, Sum C =—3.70; “Profit“ group I 2 0.75, Sum C =-3.38). Phillips and Smith ( 47 ), consi- der persons who give more total C than M to be impulsive, that they “do not carefully consider decisions and react without adequate evaluation, judgement or inhibition.“ More specific to our findings in this respect we note some of the statements of Beck ( 5 ), as regards the occurrence of high white space count (he considers 10% white space in a record to be “significantly high“) along with a preponderance of CF and C and an extratensive experience balance. The “Passion“ group produced only an average of about 7% S but the “Profit“ group did produce over 10% 8. Beck ( 5 ), states in this re- gard that “when 8% is high in an extratensive eXperience bal- ance, the person is having difficulty in controlling his 1m- pulses ....there is a trend to discharge them in an imma- ture, undisciplined pattern....aggressions find their way out in ways likely to be unpleasant to others in contact with the individual.“ He also associates the preponderance of color over movement with the “strong attack is the best defence“ strategy, saying that “the impulsivity and lack of restraint in the C and CF are accentuated when found in a cluster of many white spaces and only few fantasy responses.. the oppositon energy turns against the outside world and the person is not equipped to consume it from within.“ 8 (white space response) is the fourth (and last) Rorschach variable in which there was found to be a signifi- cant difference in productivity by the two murderer groups and Rapaport's “Fell-Adjusted“ Patrolman group. It is inter- esting to note here that the “Passion“ group's production of 8, although somewhat less (ave. S =-l.30 as compared to 1.95 ave. for the “Profit“ group) than the “Profit“ group, was significantly greater than that of the Patrolman group (P = .034), whereas the “Profit“ group's production of 8 fall (barely) short of being significantly greater than the Patrolman group (P =-approximately .07). As we have seen above in Beck's statements, it is usual to ascribe opposi- tional tendencies to those individuals (who are of normal intelligence and without organic pathology) who produce 8 responses. There has been contention as to the meaning of 8 however, and at least one recent investigator, Nelson 49 ( 46 ), reported finding white space responses to be unre- lated to oppositional tendencies. We have noted here that the Rorschach findings show the two murderer groups to be composed of immature, impulsive individuals who have a rather negativistic approach to their environment. They are quite egocentric, lacking both in social consciousness and the capacity for good empathetic relationships with others. They do have some ability to es- cape some of the stress in their environments through passi- vity (as measured by Y) and do maintain fairly adequate, 1.6. within normal limits, contact with reality (as measured by P and F¥%). Their main defense would be that of repression, the bulwark of the hysteric. Test productivity is constrict- ed (low R), there is a poverty of associative content (high A and F%) and a lack of mature mechanisms for delaying im- pulses (low FC and M). These are the implications of the Rorschach Test findings as they are listed in Tables III and IV. B.‘ The Minnesota Multiphasic Personality Inventory find in gs Since the M.M.P.I. is a somewhat newer and less widely 'used.instrument in clinical psychology than Rorschach's Test, .a brief note of introduction to it is advisable. According to Schofield ( 51 ), the M.M.P.I. is a “psychiatric inven- tory designed primarily to furnish an external, objective source of information bearing on the detection and diagno- sis of personality aberrations.“ A great deal of research has been directed toward the empirical validation of this instrument and it has gradually come to enjoy the status of . 3:1.) I 50 being a useful tool in clinical psychology. It has its drawbacks, the most apparent of which are its excessive num- ber of items (550), which usually take well over an hour to complete, and the fact that it is based on a psychiatric classification which is notoriously inaccurate and disputa- ble, but it is acknowledged to be the best inventory of its kind in existence. It carries four scales which are direc- ted primarily at measuring the validity of the subject's record and it measures personality deviations on nine cli- nical scales. The validity measures are the “Cannot Say“ L, F, and K scales. The clinical scales are: He - hypo- chondriasis, D - depression, Hy - hysteria, Pd - psycho- pathic deviate, Mf - masculinity—femininite, Pa - paranoia, Pt - psychasthenia, Sc - schizophrenia, and Ma hypomania. .A brief note of interpretation as given by Hathaway and Monachesi ( 31 ), for each of these scales can be found by the reader in Appendice #3 of this manuscript. It is quite apparent from an examination of the data presented in Table VI and its graphic presentation in Table [X that the murderer groups are considerably more deviant in ‘their'responses to the M.M.P.I. than Schmidt's “normals“ and also, to a much lesser extent, more deviant than the General Prisoner group. It is also apparent that, of the two murder- er groups, the “Profit“ group is generally more deviant than the “Passion“ group. Even though the murderer groups do da- viate considerably from the “Normals“ employed in this study, it rmust be noted that the majority of their scaled scores 51 (Table VI) do fall quite short of the scaled score of 70 that is quite generally regarded among M.M.P.I. users as significant of a “disabling abnormality“ in the particular area represented by the scale on which it occurs. Thus, a word of caution should be inserted at this time as to the 'interpretation of the significant differences between the M.M.P.I. scores of the groups used in this study, especial- ly as regards the “normal“ versus murderer group compari— sons. A statistically significant difference between groups on one of the M.M.P.I.‘s nine clinical scales does not in regard to the psychiatric category represented by that scale. Such a difference, of course, would imply that one of the groups possessed a significantly greater amount of the attributes represented by that particular scale. The most significant difference between the “Profit“ and “Passion“ murderer groups appeared not in a clinical scale but in the validating scale F where the “Profit“ group made a significantly higher score than did the “Pas- sion“ group (P = .01). The usual interpretation (30) of such a high scaled F score (67.0) as was obtained by the “Profit“ group, is that of persons attempting to put them- selves in a bad light, or being careless and inconsistent. in their responses, or as in indicative of “general malad- justment of a severe type.“ The last of these alternatives seem to this author to be the most plausible in the case of the “Profit“ group, especially so in view of their close to ° 52 average scores on the validating scales of L and F and their moderately high score on the Sc scale. If this group were “plus-getting“ or “faking bad“, one would expect a drop in their K and L scores and the F and So scales have common bizarreness about them that have led some investi- gators to regard a moderately high F score (along with a high So score) to be crucial to the diagnosis of Schizo- phrenia with the M.M.P.I. Another significant difference found between these two groups was on the validating scale K. Here the “Passion“ group made a significantly higher score than did the “Profit“ group. Hathaway and Mckinley ( 30 ), and Hathaway and Monachesi ( 31 ), consider higher K scores to indicate defensiveness (not necessarily a con- scious defensiveness, however). The only significant dif- ference between the “Passion“ group and “Profit“ groups on any of the clinical scales was in terms of social inter— action (Si). The “Profit“ group's scaled score here was only 55.0, but this was significantly greater (P'2 .05) 'than the “Passion“ group's 81 score and the General Pri- sonar group's Si score as well (no 81 score was given for Schsudt's “normals“). We can safely say that the “Profit“ group tends to hold back more in personal interrelation- ships and to be more socially seclusive than the “Passion“ and the General Prisoner groups. The finding of a “moderately high“ Mf score of 59.3 in the case of the “Profit“ group, whereas the “Passion“ groups If score was only 53.8, seems somewhat unusual in view of 4‘s 53 the findings of Kurland, Morgenstern and Sheets ( 41 ). They described the twelve wife-murderers that they studied as having “feminine-maternal attitudes toward their off- spring“ and strong passive-homosexual tendencies. In our study the wife-murderers (“Passion“ group) do not show any very strong feminine interests but the “Profit“ murderers do. The highly elevated Pd score obtained by both of our murderer groups (and the General Prisoner group as well), along with the concomittant finding of moderately high Ma, 50, and Pa scores, must be considered as pathognomic of 1 very strong antisocial (or at least asocial) tendencies on the part of the individuals who comprise these groups. The P1 scale has been found to be one of the most useful scales on the M.M.P.I. in terms of its high validity, and lately Hathaway and Monachesi ( 31 ), have found it to be equally useful in predicting juvenile delinquent behavior. The ab- sence of deep emotional response, inability to profit from experience and disregard for social mores implied by the high Pd scores are here given added impetus by the tenden- cies toward disregard of social convention, weak reality testing, and heightened suspiciousness implied by the “mo- derately high“ scores on the Me, So, and Pa scales respect- ively. The greatest amount of abnormality or deviation in this respect is in the “Profit“ murder group, whose scores on these scales (Pd = 74.2, Ma : 62.6, Sc =— 62.7, and Pa = 59.4) are particularly deviant. The General Prisoner 54 and “Passion“ groups are about equally deviant in this re- spect, with no significant differences between any of the three groups on the Pd, Ma, Sc or Pa scales even though there was a 5.8 point difference between the “Profit“ group's Pd score of 74.3 and the “Passion“ group's Pd score of 68.4. The one consistently significant difference in the per- formances of the two murder groups and the General prisoner group was on the Ry scale. Both the “Passion“ group and the “Profit“ group made significantly higher scores on this scale than did the General Prisoner group, on the raw score comparisons as well as on the scaled score comparisons We can say here that the individuals in the murderer groups are more emotionally labile, psychologically immature and more likely to avoid facing critical problems than are the individuals in the General Prisoner group. The “Profit“ murderer group also made a significantly higher scaled D score than did the General Prisoner group ( P = .05 ). As is noted by Hathaway and Monachesi ( 31 ) “higher (D scale scores normally occur when the subject is in trouble“ and these men were serving a life term for first degree murder when they were given this M.M.P.I. The “Pro- fit“ group's high F score, with its implication of severe general maladjustment also proved to be statistically great- er than the General Prisoner's F score ( P 3 .01 ) and the “Passion“ group's moderately high scaled I score, with its implications of defensiveness and lack of candor also proved to be statistically larger than the General Prisoner's scaled K score ( P =-.05. The scaled scores and profile of the “Profit“ group in many ways resemble those published by Clark ( l4 ), of a group of forty-three Army general prisoners who were given the (psychiatric) diagnosis of Anti-Social Personality“. There are differences in that the F and the Hf scores are sizeably larger and the Ma score somewhat smaller for the “Profit“'group than Clark's “Anti~Social Personality“ group, but their mean scores on K, Hs, D. Hy, Pd, Pa, Pt and So are closely similar. Our “Profit“ murder group does seem to be even a bit more deviant or maladjusted than this group of Clerks. The “Profit“‘group's elevated F score, along with with its Pd score of 74.2 and its six other clinical scales ( D, Hy, If, Pa, 8c, and Ma ) that score right at or above the 60 line, gives a somewhat bizarre quality to this group's performance on the M.M.P.I. The ”passion“ group's scores are also generally elevated, but not to the same degree as those of the "Profit‘r group, and the "moderately high“‘K score of the “Passion“ group denotes perhaps a bit of neuro- tic defensiveness in contrast to the more blunt and open re- sponsivity of the “Profit“‘group. To be sure, in evaluating the performance of both of these groups on the M.M.P.I. we cannot escape the conclusion that we are dealing here with antisocial characters, but it does appear that there is a 55 56 difference in degree of antisocial tendencies between the “Profit“ and “Passion“ murderer, and possibly a basis for postulating a difference in type. VI. CONCLUSIONS In our comparisons of the two murderer groups on both their Rorschach and their M.M.P.I. results, we find a great many similarities and few statistically significant differ- g, ences. The Rorschach and the M.M.P.I.results were largely { complementary with the Rorschach results disclosing a very ; immature, egocentric and impulsive emotional structure in a K negativistic, repressed and unempathetic personality; the M.M.P.I. a shallow, antisocially maladjusted personality that has little capacity to profit from experience. As re- ported by Lindner ( 48 ), the sign approach to the Rorschach has little value in disclosing the presence of the psycho- pathic“ personality structure, even when it shows up strong- ly on the M.M.P.I.. To merely relegate these two groups of murderers to the wide and relatively undefined statue of “psychopathic perso- nalities“ and attribute the obtained test differences to dif- ferences in degree of psychopathic deviance would seem to _ this author to not only waste the obtained data but to be an error in clinical judgement as well. If‘we consider, as does Gilbert ( 26 ), that the actual mental deviation invol- ‘ved in psychopathic behavior may range through all types of ‘neurosis and even psychotic conditions, we can well approach 57 the Problem of classifying these two murderer groups as sep- arate, though similar, personality types. The “Passion“ murderers, with their extremely immature, labile emotional reactivity, inability to form good interpersonal relation— ships, strongly repressive, quite neurotic-type defenses, and a marked tendency toward antisocial behavior, presents marked characteristics of both the neurotic hysteric and the a antisocial 'psychopath“. The “Profit“ murderers, also are extremely immature psychologically and emotionally labile but show a lesser tendency for emotional reactivity; they : are not only unable to form good interpersonal rslation~ y ships, but avoid close contact with others, being asocial as well as strongly antisocial. Their defenses include not only repression but withdrawal and avoidance as well, with the resulting implication being the type of personality that Gilbert ( 86 ), refers to as a “schizoid psychopath.“ in view of the exploratory nature of this study, it re- mains for further investigations to support or alter the a- bove implications. A content analysis of the Rorschach pro- tocols of the two murderer groups used in this study might prove helpful in this respect, as might the obtaining of similar data on convicted murderers in other regions. Further research in this area could also well include a comparison of the personalities of a group of individuals ‘who have been convicted of armed robbery with a group who have been convicted of murdering while in the act of robbery 58 (our “Profit“ murder group), a comparison of wife-deserters and wife murderers (our “Passion“ murder group), and a com- parison of a group of convicted murderers with others con- victed of crimes against prOperty. VII. SUMMARY An exploratory personality study of forty convicted and imprisoned murderers was conducted by the use of the psycholOgical tools of the Rorschach Test and the Minne- sota Multiphasic Personality Inventory. These forty mur- derers consisted of a group of twenty men who had been con- [m .-).-s'aa nib-afifl-Tw 4".- ' victed of murdering their wives (this group was designated as “Passion“ murderers) and a group of twenty men who had been convicted of murdering while committing armed robbery (this group was designated as “Profit“ murderers). Obtained results showed both murderer groups' perfor- mance to deviate considerably from that of the non-prisoner groups, to a much lesser extent from that of the prisoner groups and even in some aspects from each other, with the “Profit“ murderer group being more deviant than the “Passion“ murder group. Analysis of the findings of these M.M.P.I. and the Rorschach findings reveal the “Passion“ murderers to be extremely immature, emotionally labile, easily disturb- ed and unable to form good interpersonal relationships, to utilize strongly repressive, neurotic type defenses, and to show a marked tendency toward antisocial behavior. The 59 “Profit“ murderers were found to be extremely immature psy- chologically and emotionally labile but did show a lesser tendency for emotional reaction than did the “Passion“ group. Furthermore, the “Profit“ group was found not only to be unable to form good interpersonal relationships, but to avoid close contact with others; showing asocial as well as strong antisocial tendencies. The obtained results lead to the implication of the presence of psychopathic personality types such as have been proposed by Gilbert ( 26 ), with the “Passion“ murderer re- sembling the “hysterical psychopath“ and the “Profit“ murder- er the “schizoid psychopath.“ Further evidence is needed in this area and it was suggested that analysis of similar data obtained from convicted murderers in other regions of this country would be helpful in evaluating the personality structures of murderers and that data from the members of other criminal categories, such as armed robbery, be evalu- ated and compared with that of murderers. Appendice #1 General Instructions to Subjects I am conducting a study of men who are convicted and imprisoned for first and second degree murder. You men were put on call by me because you are now doing time for either first or second degree and I would like to get your cooper- ation in this study. All that will be required of you is 5 that you take two tests, one of which is a paper and pencil i test that will probably take you a couple of hours to fin- ish, and the other one is a test that I have to administer to you individually, which will probably take only about an hour or so. You are not required to take these test, if you don't want to take them there will be nothing further said about it. The prison officials are not concerned with who does or doesn't cooperate. This is a personal research on my part and part of my required schoolwork. You have nothing to lose but a few hours of your time by taking them and, for that matter, quite probably nothing to gain. It is the find— ings as a whole that are of primary interest, and individual test results will be kept strictly confidential in that quite probably no one other than me or my thesis adviser will ever know exactly who was tested and how each individual's tests turned out. Fern .4. . n .i. .I a" t!) illel Appendice #3 A Brief Index to the Scales of The M.M.P.I., by Hathaway and lonachesi (31) THE ? SCALE This score is obtained by counting the number of items in the Inventory that the subject did not answer (those he placed in the category Cannot Say). A high score would mean that, because so many items were left unanswered, no conclusions would be safe. Subjects are routinely urged to a say True or False to nearly all the items. “‘15.. vi THE L SCORE This little group of items yields a score tending to indicate any naive attempt on the part of the subject to put himself in a good light, chiefly with reference to per- sonal ethics and social behavior. High scores are obtained on persons who try (often unconsciously) to answer all the items in ways that will seem to fit most clearly into the subject's interpretation of the moral code regardless of secret kDOWIedge about himself to the contrary. Such at- tempts could be called defensiveness, or in more flagrant cases, “faking good.“ ...Ar‘_~r.~:.-l ._ .4 . 1~'- ... ..1 THE F SCORE The F score is somewhat the opposite of the L score. Persons obtaining a high F score often seem to be attempt- ing (perhaps unconsciously) to show themselves in a bad light. They may be “faking bad.“ Sometimes such persons are merely overly candid. This is called “plus getting“ in M.M.P.I. jargon. The F score is also high if the sub— ject, for any reason, fails to answer carefully or con- sistently; that is, one will obtain a high score for per- sons who cannot read well enough to make discriminative re- sponses and, what is more significant for high school use of the Inventory, this scale will be high when the student answers carelessly, making random or facetious responses to the items. A third source of moderately high F scores is general maladjustment of a severe type. THE X SCORE This variable is much more complex and less obvious than the L and F scores in its import. Special interpre- tation is not usually indicated because the score is mix- 61 ed with five of the clinical scales in a way that proper- ly used the score as a correction. In general, higher K. scores, like L but in a more subtle way, indicate defen- siveness and lack of candor, and low K scores, like high F scores, like high F scores, indicate a degree of frank— ness and self-criticality. The modification of scores on the five clinical scales by use of K was shown to be justified within the borderline abnormal score range. 'Routine use of K within the normal score range was not specially validated but is usually prac- ticed to simplify application and interpretation. SI, SCALE 0 This is a measure related to social introversion and extroversion. Persons scoring high on the scale tend to hold back in personal inter-relationships and are likely to be socially seclusive. When the score is low, the individ- ual is more extroverted, tends to join organizations with a primarily social purpose and to take an active part in them. HS» SCALE 1 This scale was derived by comparison of normal persons to patients having many physical complaints with a prepon- derantly psychological basis. These complaints can be in part an outcome of obvious tissue pathology but are usually so varied and symbolic in nature that they are clinically . classed with what is currently termed psychosomatic illness. D, SCALE 2 This scale was derived from persons who were depressed. Individuals obtaining a high score on it feel unsure of them- selves and of the future and often they are sad and blue. Higher scores normally occur when the subject is in trouble and in a sense the absence of a higher score from a person in trouble is an unexpected sign since he seems not to be responding in the modal way. HY, SCALE 3 This scale is closely allied to scale 1, but it is evi- dence of more complete symbolic elaboration of the physical symptoms. The elaboration usually takes the form of cultur- ally respectable psychosomatic syndromes such as allergy or functional cardiovascular disturbances. The person with a. 'high score on scale 3 defends the psychological elaborations against recognition of their neurotic character by himself or others. Persons who have high.scores are more likely to avoid or run away from crises, than to face them and.work in— to them directly. In their avoidance of crises, they tend to try to adjust by “buck up“'attitudes and when they are 62 pushed too far, they develop physical or even psychologi— cal illness as a sort of solution. PD, SCALE 4 The syndrome of symptoms that were characteristic of the patients from whom this important scale was derived is indicated by the diagnostic term_psychopathic deviate. Of- ten young and delinquent, these individuals always impress the clinician with their failure to be controlled by the ordinary mores of society. They seem little affected by remorse and do not appear to be particularly modified by censure or punishment. They are likely to commit asocial acts, but these frequently lack obvious motive. It should not be assumed that all delinquents will score high on scale 4. The syndrome requires more than mere delinquency or criminality to establish the diagnosis. One may only be confident that any delinquent or criminal group will include an unknown number of persons who, if studied more broadly, would be so diagnosed. Youngsters who score high on scale 4 are more likely than adults to be in conflict with their families, but the social conflict can, of course, be more extensive. MF, SCALE 5 This is a measure of masculinity or femininity of in— terests. In males, high scores are indicative of general feminine interests as these appear in contrast to the aver- age male; in females, high scores indicate masculine inter- ests. Persons who have homosexual tendencies often earn high scores; particularly is this true of males. High scores are not, however, safely interpreted in this way. This scale could not be scored on the early forms of the M.M.P.I. and the earlier papers do not use it. PA, SCALE 6 This scale is a measure of undue interpersonal sensi- tivity; at its extreme this may be a paranoid feeling about other people in which the subject feels mistreated or threatened. Probationers and other subjects under close social control are.likely to show a moderate elevation of this score. PT, SCALE 7 This scale is related to compulsions and obsessions. IPersons with high scores are in some ways excessively me- ‘ticulous or overly conscientious. This fussiness rarely characterizes all phases of a person's behavior, but tends toi‘be most noticeable in restricted aspects. Others of ‘Ehose obtaining high scores worry more than is justified, Winile still others are overly introspective and self-cri- tical. Some persons who senselessly repeat asocial acts 63 . s o v. (9:53 a. are driven by compulsive energy. Arsonists, exhibitionists, and rapists are extreme cases likely to show this syndrome. SC, SCALE 8 This is a measure related to the degree to which the person thinks and reacts like others about him. At the ex- tremes, or with certain bizarre symptoms, the scale is an indicator of the mental disorder schizophrenia. More gen- erally, the scale is a measure of the way in which the person may distort some aspects of the world about him, perceiving it differently than others and reacting to it in unusual ways. MA, SCALE 9 t This is a measure related to enthusiasm and energy. Persons scoring high on the scale become readily interest- ed in things and approach problems with animation. When this becomes abnormal, the activity may lead to antisocial acts or to irrational manic behavior. Young people are normally characterized by a considerable amount of the fac- tor this scale measures. When they have too much of it, they DeCOme restless and frequently stir up excitement for ex- citement's sake alone. I r‘rr-rr --r ‘f‘ . In addition to these, there are a number of other scales that have been developed to be scored on the M.M.P.I. Undoubtedly some of the others will, in time, be as helpful as those outlined here. In the present collection, however, we shall restrict ourselves to findings with the scales described above. 64 BIBLIOGRAPHY Alexander, F. and Staub, H. The Criminal, The Judge, and the Public, New York, The MacMillan Company, 1931. Altus, W.D. “Some correlates of the group Rorschach and the schizophrenic scale of the group M.M.P.I. among two groups of 'normal' college students.“ J. Consult. Psych. 1948, 12, 375-378. Banay, R.S. “Study in Murder“, The Annals, Vol.284, -~ November, 1952, pp. 26-34. Beck, S.J. Rorschach's Test. Vol. II. A variety of personality pictures. New York, Grune and Stratton, 1945. Beck, S.J. Rorschach“s Test. Vol. III. Advance; in Interpretatign, New York, Grune and Strat- ton, 1952. Beck, S.J., Rabin, A.I., Thiesen, W.G., Herman and Thetford, W.N. The normal personality as projected into the Rorschach Test. J. Psychology, 1950, 30, 241-298. Berg, I.A. and Fox, V. “Factors in Homicides Committed by 200 Males”, J. Soc. Psychol., 1947, 26, PP. 109-119. 65 10. 11. 12. 13. 14. 15» 16» Bochner and Halpern F. The Clinical Application of the Rorschach Test. New York, Grune and Stratton, 1945. Brasol, B. The Elements of Crime, New York, Oxford University Press, 1927. Brearley, H.C. Homicide in the United States, Chapel Hill, Univ. of North Carolina Press, 1932. Brearley, H.C. The Negro and Homicide, Social Forces, Cassity, J.H. “Personality Study of 220 Murderers“, J. Crim. Psychopathol. 1941, 2, 296-304. Clark, J.H. “Some M.M.P.I. correlates of color re- sponses in the group Rorschach.“ J. Consult. Psych. 1948, 13, 384-386 Clark, J.H. The relationship between M.M.P.I. scores and psychiatric classification of Army general prisoners. J. Clin. Psychol., 1952, 8, 86-89. Cochran, W.G. and Cox, G.H. Experimental Designs, New York, Wiley, 1950. Druvant, B. and Waldrop, F. “The Murderer in the Men- tal Institution“, The Annals, Vol. 284, Nov. 1952. PP. 35-44 66 — A -p‘ 17. 18. 19. 20. 31. 83. 23m 234. Dublin, L. and Bunzel, B. Thou Shalt Not Kill, Sur. Graph., 1935, 24, pp. 127-131. Fabricant, B. Suggestibility and the Rorschach, J. Clin. Psychol., 1955, 11, 309-311. ‘ Ferri, E. Criminal Sociology, Modern Criminological Science Series, Boston, Little, Brown and Co. 1914. Fisler, D.W. and Baughman, E.E. Relationships between Rorschach scoring categories and the total number of responses. J. Abnorm. Soc. Psychol., -w» 1953, 17, 85-33. Frankal, E. “One Thousand Murderers“; J. Crim. Law and Criminology, Vol. 29, May 1938 - April 1939, DD. 672-688. Frankle, A.H. Rorschach human movement and human content responses as indices of the adequacy of inter- personal relationships of social work students. Unpublished Ph.d. thesis, Uhiv. of Chicago, 1953. Garofalo, R. Criminology, translated from the Italian by Robert W. Miller, Little, Brown and Company, 1914. Gillin, J.L. The Wisconsin Prisoner, University of Wise. press., Madison, Wisconsin, 1946. 67 25. 26'. 27K. 28. 29. 30. 31. 38. 33. Gillin, J.L. “Murder as a Sociological Phenomenon“, Vol. 284, Nov., l952, pp. 20-25. Gilbert, G. The Psychology of Dictatorship. Ronald Press 00., 1950. Goodman, L.A. Kolmogorov-Smirnov tests for psycho- legical research. Psychol. Bull., 1954, 51, 160-169. Guilfcr, J.P. Fundamental Statistics in Psychology and Educationy_New York, McGraw-Hill Book Co., Inc., 1950. _ V‘—--v—u-1F"N4‘.J‘ .. ... ,I-' e .....‘ a .A..i A ~ . Harlan, H. “500 Homicides“, J. Crim. Law and Crimi- nology, 40, 1950, pp. 736-752. Hathaway, S.R. and McKinley. J.C. Manual for the MLM.P,I,, New York, Psychol. Corp., 1943 Hathaway, S.R. and Monachesi. Analyzing and Predicting Juvenile Delinguency with the M.M.PLI. 1953, Colwell Press, Inc., Minneapolis. Henderson, D.K. “Psychopathic Constitution and Criminal Behavior“ in Craig, East, et. a1. Mental Ab- normality and Crime. MacMillan and Co., Ltd., London, 1949. Hodgdon, L.L. Social Characteristics of 400 First De- gree Prisoners in Michigan Prisons. M.A. the- sis, Michigan State College, 1946. 68 34. 35. 36. 37. 38. 39. 40. 41. Hooton, E. The American Criminal, Cambridge, Mass., Harvard University Press, 1939. Karpman, Ben. “A Psychoanalytic Study of A Case of Murder“, Psychoanalytic vaiew, 1951, 38, pp. 139-157 and 245-270. Karpman, Ben. “Criminal Psychodynamice: A Platform.“ Archives of Criminal Psychodynamics, Vol. I, #1, 1955, 3-96. King, J. A Theoretical and Experimental Consideration of the Rorschach Movement Response: Its rela- tion to the neuropsychiatric patients orienta- tion to his problem. Klopfer, B., Ainsworth Mary D., Klopfer, W.G., and Holt, S.R. Developments in the Rorschach Technique: Vol. I;_Technique and Theory, Yonkers, World Book Co., 1954. Klopfer B., and Kelly, D., The Rgrschach Tgchnigue. Yonkers, World Book Co., 1946. Knopf. I.J. Rorschach summary scores in differential diagnosis. J. Consult. Psychol., 1956, 20, Kurland, A., Morganstern, J., and Sheets, Carolyn. “A comparative study of wife murderers.“ The J. of Soc. Therapy, Vol. I, No. 2, Jan, 1955, 12-26. 69 43. 43. 44. 46. 47. 49. Lindner, R.M. The Rorschach Test and the diagnosis of Psychopathic Personality. J. Crim. Psycho- pathol., 1943, 5, pp. 69-93. Massey, F.J., Jr. The distribution of the maximum deviation between two sample cumulative step- function. Ann. Math. Statist., 1951, 22, 125-128. McCall, R.J. and Doleys, E.M., Jr. Popular responses on the Rorschach Test in relation to the num- —uv-‘a.aa.-?nz 9.1". ’2.- . x _ a . . l . - . ‘ , ber of responses. J. Clin. Psychol., 1955, 11, 300-303. Metropolitan Life Ins. Co. Statistical Bulletin, 1955, Vol. 36, Nos. 4 and 6. Nelson, W.D. An Evaluation of the White Space Response on the Rorschach as Figure-Ground Reversal and Intellectual Opposition. Unpublished Ph.D. thesis, Mich. State Univ, 1955. Phillips, L., and Smith, J.C. Rprschach Interpretation! Advanced Technigue, New York, Grune and Strat- ton, 19530 Rapaport, D. Diagpostic Psychglogical Testing, Vols. I and II, Year Book Publishers, Chicago, 1946. Reik, T. The Unknown Murderer, London, The Institute of Psycho-analysis, 1936. 70 50. 51. 52. 53. 54. 55. 56. 57. 58. Schachtel, E.G. “Projection and its relation to char- acter, attitudes and creativity in the kines- thetic response.“ Psychiatry, 1950, 13, 257-273. Schofield, W. Changes in responses to the M.M.P.I. following certain therapies. Psychol. Monog., 1950, Vol. 64, No. 5. Selling, L.S. An Analysis of the Causes of Homicidal Mortality. Ph.D. thesis, University of Michi- gan, January, 1940. Smith, M.H. The Psychology of the Criminal. New York, Robert McBride and Company, 1923. The Annals of the American Academy of Political and Social Science. 1952, Vol. 284. Thompson, C. “M.M.P.I. correlates of certain movement responses in the group Rorschach's of two col- lege samples.“ J. Consult. Psych., 1948, 12, 279-383. Vold, 6.8. Extent and Trend of Capital Crimes in the United States, The Annals, Nov. 1952. White, W.A. Crime and Criminals, New York, Farrar and Rinehart, 1933. Whitfield, R.N. The Homicide Situation in the United States. The American J. of Public Health, Vol. 27, Oct. 1937. 71 lm.-r.u".- r A4A<4U_ . --C.. ‘4“..1‘5 ' Hardin fines] Demco-293 "7'1? 11'1le?”11111117111131?