Remorse and Psychopathy
Psychopaths are generally seen as lacking the capacity for remorse (Hare 1975), although the studies of Foulds, Caine, and Creasy (1960), Marks (1965), and Gudjonsson and Roberts (1983) indicate that psychopaths report experiencing strong feelings of remorse after transgression. An important distinction here is the primary--secondary categorization of psychopathy. In contrast to primary psychopaths, secondary psychopaths exhibit strong symptoms of anxiety and autonomic arousal (Lykken 1957).
Mealy (1995) discusses some important differences between primary and secondary psychopaths, which are relevant to feelings of guilt. The main difference is that secondary psychopaths are primarily antisocial due to exposure to environmental risk factors, whereas the antisocial behaviour in primary psychopaths is primarily determined by their genotype (i.e. have a substantial genetic component). Another difference between the two types of psychopaths is that secondary psychopaths, in contrast to primary psychopaths, are capable of experiencing some sincere social emotions (e.g. guilt, shame, sympathy, empathy).
The work of Gudjonsson and Roberts (1983) into feelings of guilt and self-concept in secondary psychopaths raises some important questions about the nature of guilt. Twenty-five male and 25 female psychopaths, who were being treated in a therapeutic community in England, were compared with normal subjects with regard to Morality-Conscience Guilt as measured by the Mosher True-False Inventory (Mosher 1966). The psychopaths scored significantly higher on the guilt inventory than the normal subjects, which contradicts the common view that psychopaths experience little guilt after transgression. The Semantic Differential technique (Osgood, Suci and Tannebaum 1957) was administered to the patients in order to measure their self-concept, including feelings of guilt and shame when transgressing. The patients rated themselves on the ten bipolar dimensions with regard to the following concepts: 'Myself as I am', 'Myself as I would like to be', 'Myself when I lie', 'Myself if I were to steal', 'People who lie', and 'People who steal'. Factor analysis of the ten bipolar dimensions revealed three factors, referred to as an 'evaluative' (good--bad), 'potency' (strong--weak), and 'guilt' (remorseful--unremorseful, and ashamed--unashamed) factors, respectively.
The psychopaths had significantly greater semantic distance between their self and ideal self than the normal subjects on all three factors, which reflects their poorer self-evaluation. As far as the guilt factor was concerned, the normal subjects had very little discrepancy between their self and ideal self, whereas the psychopaths wished to feel less guilt and shame. The normal subjects reported a marked increase in their feelings of guilt and shame when they were transgressing (i.e. lying or stealing), whereas in the case of the psychopaths there was no change in their degree fo guilty feeling when transgressing.
The findings suggest that some psychopaths have poor self-concept which is reflected in internal distress and negative preoccupations, which they label as 'guilt' or 'shame', regardless of whether or not they are engaged in antisocial behavior. Normal subjects, in contrast, only experience feelings of guilt when they perceive violation of some norms. If some psychopaths do experience strong feelings of guilt, which are unrelated to specific situational transgression, then this may explain why their feelings of guilt fail to inhibit antisocial behavior. That is, engaging in antisocial acts does not make them feel any worse than they already feel.
If the guilt reported by some psychopaths represents genuine feelings of guilt, rather than mislabeling, how does one interpret their apparently high degree of guilt? One possible explanation may relate to these psychopaths being punished indiscriminately in childhood for both prosocial and antisocial behavior. Irrespective of the moral value of their behaviour, they are punished. They can never do anything right in the eyes of significant person in their lives. They consequently develop a conditioned response to their own behaviour which becomes generalized rather than being situation specific to legitimate transgression. Placing this finding within the framework of Solomon and his colleagues in to the differences between the development of feelings of guilt and resistance to temptation (Solomon, Turner and Lessac 1968), the psychopaths in the Gudjonsson and Roberts (1983) study never learned to resist temptation due to the timing of the punishment implemented in childhood, which resulted in strong indiscriminate emotional responses to perceived punishment.
The psychopaths in the Gudjonsson and Roberts study are best described as secondary psychopaths in the view of their high level of trait anxiety and physiological reactivity (Gudjonsson and Roberts 1985). It would be interesting to repeat the Gudjonnson and Roberts (1983) study on primary psychopaths in order to see if differences in guilt response can be documented experimentally between primary and secondary psychopaths.
An interesting question is whether under certain circumstances guilt may actually be a precipitating factor in the commission of a crime. In other words, rather than guilt preventing a crime can it actually increase the likelihood of its occurrence? The empirical evidence for this proposition is lacking, but this does not mean it could not be true. Freud (1957) put forward the proposition that this can happen and referred to such cases as 'criminals from a sense of guilt'. According to Freud, here guilt-ridden individuals are preoccupied by their free-floating guilt, like the secondary psychopaths in the Gudjonsson and Roberts (1983) study, and by committing a crime the guilt is attached to something tangible and this is accompanied by a sense of mental relief. Another possibility is that a strong feeling of free-floating guilt is associated with emotional instability and strong autonomic arousal, which can lead to impulsive and antisocial behaviour by virtue of its drive propensities such as Eysenck originally proposed for neuroticism (Gudjonsson 1997). Feelings of guilt can be pathological, either because they are not attached to a specific act of transgression (free-floating) or because they interfere with the effective prosocial functioning of the individual.
* Cox, Murray. Remorse and Reparation. United Kingdom: Jessica Kingsley Publishers Ltd, 1999. 91-93.