Research Methods in Psychology

Coping with guilt and shame: a

narrative approach

Mia Silfver *

University of Helsinki, Finland

Autobiographical narratives (N597) of guilt and shame experiences were analysed to determine

how the nature of emotion and context relate to ways of coping in such situations. The coding

categories were created by content analysis, and the connections between categories were analysed

with optimal scaling and log-linear analysis. Two theoretical perspectives were tested: the view that

shame generally is a more maladaptive emotion than guilt, and the view that in situations where

responsibility is ambiguous, both guilt and shame feelings are likely to be maladaptive. In line with

the latter, chronic rumination was more likely to occur in situations where responsibility was

ambiguous compared to situations where the respondent’s responsibility was clear, regardless of

emotion. In addition, reparative behaviour was less frequently reported in shame situations than in

situations where the respondent felt guilty or both guilty and ashamed. The findings supported the

view that the nature of emotional reaction and the nature of the situation both affect the ways of

coping.

Introduction

Guilt and shame are often seen as emotions that have an important role in moral

behaviour. Some researchers emphasise that guilt especially motivates moral

behaviour (Baumeister et al., 1995; Hoffman, 2000), while others believe that both

guilt and shame can motivate moral behaviour, depending on the situation (Lindsay-

Hartz et al., 1995). Some believe that shame has only negative implications,

regardless of the situation (Tangney & Dearing, 2002). When defining guilt and

shame, different researchers have emphasised different aspects. Within the

psychology of emotion, shame has often been included in the list of primary

emotions, because it is characterised by certain gestures and expressions, whereas

clear distinctive features have not been found for guilt (see Izard, 1971).

Based on several empirical studies conducted in the USA, Tangney (1995) has

identified eight dimensions on which guilt and shame differ. First, shame is directed

to the global self, ‘I did that horrible thing’ whereas guilt focuses on specific

*Department of Social Psychology, P.O. Box 54, 00014, University of Helsinki, Finland.

Email: [email protected]

Journal of Moral Education

Vol. 36, No. 2, June 2007, pp. 169–183

ISSN 0305-7240 (print)/ISSN 1465-3877 (online)/07/020169-15

# 2007 Journal of Moral Education Ltd

DOI: 10.1080/03057240701325274

behaviour ‘I did that horrible thing’. In shame, the self is ‘split’ into observing and

observed selves, and the self is impaired by global devaluation, but in guilt it is not.

Shame involves mentally undoing some aspect of the self, whereas guilt involves

mentally undoing some aspect of behaviour. Furthermore, an ashamed person

experiences ‘shrinking’, feels small, worthless and powerless, whereas a guilty person

experiences tension, remorse and regret. Consequently, shame is a more painful

emotion than guilt. An ashamed person is concerned with others’ evaluation of self,

but a guilty person is concerned with one’s effect on others. Tangney believes that

these differences affect the motivation these emotions create. A person feeling guilty

is motivated to confess, apologise or repair, whereas an ashamed person feels a desire

to hide, escape or ‘strike back’, to behave aggressively towards the person inducing

shame. As a result, being prone to shame would be a maladaptive tendency, whereas

guilt proneness would enhance moral or prosocial behaviour. Based on this

differentiation, Tangney has created a scenario-based measure for guilt and shame

proneness, the Test of Self-Conscious Affect (TOSCA), in which the respondents

evaluate their likelihood to react in certain ways in a set of hypothetical situations.

The words guilt and shame are not explicitly used. Using the TOSCA, being prone

to shame has been found to relate to low self-esteem, anxiety, depression and

psychoticism (Tangney & Dearing, 2002), whereas being prone to guilt has been

connected to positive characteristics, such as anger control (Lutwak et al., 2001) and

empathy (Fontaine et al., 2001). Guilt and shame scales have some overlap, but

Tangney and Dearing (2002) emphasise that shame scores should be partialled out

from guilt, because when fused with shame, guilt can become maladaptive.

However, shame-free guilt, no matter how intense, is never maladaptive.

Tangney’s view of the motivational features related to guilt and shame has faced

some criticism. The TOSCA items connect certain emotional states to specific

behavioural reactions: the guilt items include prosocial behaviour or intentions to

behave prosocially, whereas the shame items include only avoidance (e.g. ‘I would

avoid seeing that friend for a while’) and negative self-evaluations (e.g. ‘I would feel

useless and incompetent’), but no socially desirable ways to alleviate the emotion.

Therefore, the guilt scale can be seen as a measure of prosocial behaviour rather than

guilt as an emotional state. Consistent with this view, the TOSCA guilt scale has

been found to measure a different construct than other guilt measures, for example

the Guilt Inventory and the Personal Feelings Questionnaire-2 (Ferguson &

Crowley, 1997). Tangney (1995) suggests that the differences between her own

measure and those used by others are due to the fact that most other measures do

not successfully separate guilt and shame. On the other hand, Ferguson and Crowley

(1997) point out that the measures probably tap different phases of the guilt process.

Guilt items in the TOSCA include adaptive ways to alleviate guilt, whereas other

measures describe unresolved, ongoing guilt experiences, for example, ‘I have

recently done something I deeply regret’ (state guilt in the GI). They suggest that

shame and guilt can both be maladaptive if a person is not able to resolve the

situation in a constructive manner. Ferguson and Crowley recommend that

researchers use a variety of measures of guilt, because different measures tap

170 M. Silfver

different kinds of guilt. Luyten, Fontaine, and Corveleyn (2002) came to the

conclusion that the TOSCA primarily measures mild and adaptive forms and

aspects of guilt and maladaptive aspects of shame. By conducting principal

components analyses on the TOSCA, they revealed that items that had a high

loading on the guilt factor primarily were items that referred to reparative behaviour,

while the items that had high loadings on the shame factor referred to low self-

esteem. Including only the items that had high factor loadings did not change the

correlations to other variables. It seems that reparative behaviour in the TOSCA

guilt items could mainly explain the results concerning guilt’s adaptive value

compared to shame.

Against the assumption of the TOSCA that guilt always leads to reparation,

Bybee, Merisca and Velasco (1998) have shown that the initial emotional reaction

and the means to cope with the emotion can be differentiated. Using narrative data,

they have found that reacting by feeling guilty does not necessarily lead to prosocial

or reparative behaviour, but there are different possible strategies to manage the

emotion. Consequently guilt itself is not adaptive unless the person is able to deal

with the feelings of guilt in a constructive way: to correct one’s mistakes when

possible, but not endlessly punish oneself. The reactions to guilt-producing events

can be divided into three main categories: reconciliatory reactions, intropunitive

responses and rationalisations (Bybee et al., 1998). Shifting the blame away from the

self by using rationalisations, excuses and justifications has been connected to

antisocial tendencies (Bybee et al., 1998). At the other extreme, chronic, self-

punitive feelings of guilt are associated with poor mental health (Bybee & Quiles,

1998). Reconciliatory reactions help to channel guilt into positive behaviour and

prevent guilt from becoming chronic (Bybee et al., 1998). Fontaine et al. (2006)

created a scenario-based measure in which respondents evaluated the likelihood of

certain emotions and other reactions in a set of situations. They performed

multidimensional scaling based on situation variation, and found that situations

were differentiated by two dimensions. On the first dimension, guilt and shame were

at the opposite ends of the dimension, and reparation and regret were close to guilt,

whereas embarrassment, desire to disappear and concern for one’s reputation were

close to shame. The other dimension differentiated interpersonal and intrapersonal

orientation. Interestingly, ruminative, self-oriented reactions were more closely

related to guilt than shame. There appeared to be two forms of guilt: other-oriented

guilt that motivated reparation, and self-focused, ruminative guilt. A similar

distinction was found by Quiles and Bybee (1997). They performed a factor

analysis on several guilt measures and found two factors that they called

‘predispositional guilt’ and ‘chronic guilt’. Items in the factor of predispositional

guilt described guilt reactions in specific situations whereas the chronic guilt factor

consisted of items describing ongoing, unresolved feelings of guilt. Chronic guilt was

associated with different psychopathological symptoms, but predispositional guilt

was not.

Tangney and Dearing (2002) suggest that chronic guilt is maladaptive only

because it includes shame. Tangney has also tried to create a scale for chronic,

Coping with guilt and shame 171

ruminative guilt in the TOSCA, but she found it to be so strongly related to the

shame scale that it would not give any new information. One could still ask whether

chronic guilt includes shame because the person has not found a way to alleviate

guilt, and whether the actual cause of chronic guilt would be lack of coping skills

instead of shame proneness. Using the term guilt only of non-chronic, shame-free

guilt is problematic, because the term is easily misunderstood; this use is

contradictory to the research tradition in psychology and psychiatry as well as the

common understanding of the concept.

The situational context for feeling guilt or shame should be considered, because

there is evidence that in some contexts both guilt and shame can be equally

maladaptive. Ferguson and Stegge (1998) point out that many of the situations used

to assess guilt are those in which the consensual response would probably be guilt.

As a result, these measures may be unable to detect maladaptive forms of guilt:

exaggerated sense of responsibility and tendency to feel guilty in situations in which

most people would not. Ferguson, Stegge, Eyre, Vollmer and Ashbaker (2000)

tested the influence of the context on guilt in children. They found that the tendency

to feel guilty in ambiguous situations (the person has no control over the situation

and thus cannot be held responsible) was as strongly related to psychological

symptoms (anxiety, depression, externalizing) as the tendency to feel shame. The

tendency to feel guilty in ambiguous situations easily leads to chronic guilt, because

it is difficult to find a way to resolve the situation. However, Hoffman (2000)

believes that guilt in ambiguous situations (which he labels as ‘virtual guilt’) is

equally a motive for prosocial behaviour, because even if the person is not

responsible for the situation, it is usually possible to do something to improve things.

It is possible that guilt in ambiguous situations is related to both psychological

symptoms and reparative behaviour, because a person can assume so much

responsibility that it becomes unhealthy. According to Tangney and Dearing (2002),

guilt is adaptive even in situations when there is no means to correct the situation,

because one can always decide to behave differently when faced with a similar

situation in the future. But what if you are not sure what you should have done or

how you could have affected the situation? It could be assumed that guilt is most

adaptive in situations when a person knows what should be done to set things right.

The question then is: Does the person have the motivation to do it?

There is also some empirical evidence against the view that only shame-free guilt is

likely to be adaptive and the presence of shame makes prosocial behaviour less likely.

Harris (2003) studied drunk-driving offenders using his own measure based on the

definitions of guilt, shame and embarrassment found in literature. He found three

factors in his data: shame-guilt, embarrassment–exposure and unresolved shame. A

factor of shame-free guilt could not be identified. Shame-guilt was strongly related to

empathy and negatively related to anger/hostility, whereas unresolved shame was

strongly related to anger/hostility, but the relation to empathy was weak. Harris’s

data, then, suggest that regret and concern for the victim can be combined with

negative self-evaluations and fear of judgement by others and still motivate

reparative behaviour. Shame seems to be maladaptive when it does not include

172 M. Silfver

guilt but instead is combined with negative defences. Ahmed and Braithwaite (2004)

studied school bullying and found that shame displacement (managing shame by

using defences such as other-directed blame and anger) in bullying situations

appeared to increase bullying, whereas shame acknowledgement decreased it.

Shame management also partially mediated the effects of family, school and

personality variables on bullying. Shame was defined consistent with Tangney’s

(1995) criteria, except that defensive behaviour was not assumed to be always a part

of the experience of shame. It is likely that the inability to cope with shame in a

constructive way, rather than shame itself, causes problems.

Some researchers believe that both guilt and shame can be adaptive or

maladaptive depending on the situation. Lindsay-Hartz, de Rivera and Mascolo

(1995) emphasise that the adaptive value of a certain emotion can be determined by

looking at the function the emotion serves in a specific context. In some cases guilt

can serve as a defence in a maladaptive way. A person can feel guilty about

uncontrollable events, because he or she cannot accept the fact that there are

unwanted, uncontrollable events, e.g. illnesses. Feeling guilty instead of ashamed

can also be a way to avoid exploring one’s motives, because guilt only concentrates

on certain behaviour, not the whole self. Shame can help an individual to pay

attention to more permanent qualities of the self and motivate pursuit of the ideal

self. On the other hand, shame is likely to be maladaptive when the ideal self is

unattainable or unrealistic, or if the whole self is condemned on a basis of a single

characteristic, which can make a person feel helpless and unable to make changes.

Also, Barrett (1995) believes that both guilt and shame serve important, but

different, functions. Both shame and guilt highlight particular social standards for an

individual and help him or her to acquire knowledge about the self. Shame draws

attention to the self as an object perceived by others; it communicates deference and

submission to others and helps to maintain social hierarchies. Guilt, on the other

hand, helps to understand the self as an agent, brings a person closer to others and

motivates him or her to repair the harm that has been caused.

It is also worth recognizing that the research on coping with shame or guilt has

been largely conducted within individualistic cultures, and shame particularly may

be experienced differently in collectivistic cultures. Kitayama, Markus and

Matsumoto (1995) suggest that the link between shame and defensive reactions,

like anger, is typical of individualistic cultures like that in the USA, where the sense

of self as independent is valued, and shame can be interpreted as a sign of weakness.

From this perspective, hiding shame with anger is a reasonable thing to do, but in

collectivistic cultures, in which interdependence is valued, defending the self against

shame in this way is probably seen as useless. Demonstrating shame to others is seen

as better, because it helps to maintain relationships. Bagozzi, Verbeke and Gavino

(2003) compared Dutch and Filipino salespersons’ experiences of shame as a

consequence of customer actions. The actual experience of shame was very similar

for both groups (painful, self-focused emotion, the core self was felt to be

threatened), but the behavioural reactions to it were different: for Filipino employees

shame enhanced customer relationship-building, whereas for Dutch employees

Coping with guilt and shame 173

shame diminished it. Fischer, Manstead and Mosquera (1999) compared Spanish

and Dutch students and found that Spanish students were more likely to report

sharing shame experiences with others, and they also expressed more positive beliefs

about shame, for example that shame is a sign of strength and it makes others see

you positively. Shame also appeared to be a more social experience for the Spanish:

they were more likely to report shame experiences involving public performance and

social judgement, whereas the Dutch were more likely to report self-centred

experiences of personal failure (Mosquera et al., 2000). Walbott and Scherer (1995)

have obtained extensive empirical evidence of cultural differences in the experience

of shame. In collectivistic cultures, shame was found to be a rather acute, short-lived

emotional experience compared to individualistic cultures. For guilt experiences, on

the other hand, clear cultural differences were not found. In sum, coping with shame

appears to be related to cultural context.

The research question and hypotheses

In this study, narrative descriptions of guilt and shame experiences were used to

investigate how emotional experience and situational context affect coping.

According to Tangney and Dearing (2002), shame-free guilt should be more

adaptive (include more reparation and fewer defences and chronic rumination) than

shame or a combination of guilt and shame. However, it is also possible that only

guilt-free shame is maladaptive, as indicated in Harris’ (2003) study. Furthermore,

ambiguous context may be connected to more chronic rumination, regardless of

emotion, as suggested in the study of Ferguson et al. (2000).

Alternative hypotheses were tested.

1. In line with the theory of Tangney and Dearing (2002), it was expected that

reparative behaviour would be more likely, and rumination and defences less

likely, in shame-free guilt situations than in situations of pure shame or combined

guilt and shame, regardless of situational context.

2. Consistent with Ferguson et al. (2000), it was expected that chronic rumination

would be more likely to occur in ambiguous situations than unambiguous

situations, regardless of emotion.

Method

Participants and procedure

The participants were university students or students from social psychology courses in

the open university (a school where it is possible to attend university courses, but not to

get a degree). The participants were contacted through student mailing lists or course

websites, and they were provided with a link to an electronic questionnaire, where they

were asked to describe a real-life experience of guilt, shame or both emotions

simultaneously. The participation was voluntary and no payment was given, so only

those who were especially motivated to share their experiences answered, and therefore

174 M. Silfver

the sample is not representative of the students who participated on the courses or were

on the mailing lists. There were 97 participants (12 men, 85 women), who provided

from one to three narratives each. The total number of narratives was 120. The

participants’ ages ranged from 20 to 57, and the mean age was 29 (SD 8 years).

They were requested to answer the following questions:

What was the situation in which you felt guilt, shame, or both? What did you think, feel

and do in the situation?

What kind of thoughts or behaviour did you use to alleviate guilt or shame?

Were you successful in alleviating these emotions or did you continue to suffer from

guilt or shame?

Participants were also asked to describe a situation when they had done something

because they anticipated feeling guilt or shame afterwards if they did not do it.

The data was content analysed following the example of Baumeister, Stillwell, and

Wotman (1990). A deductive research strategy was applied (Mason, 2002).

Hypotheses were generated in advance, and the basic dimensions that were studied

were based on theory. The dimensions were emotion, situation, responsibility and

coping. Each story provided by participants was classified according to its emotional

content, according to whether it included guilt, shame or both. The emotion was

coded based on the respondent’s own report of whether he or she experienced guilt,

shame or both during the episode. The respondents used the terms guilt and shame

in a way which was very consistent with Tangney’s (1995) definition of these

emotions, with shame referring to self and guilt referring to behaviour. The

situations were also classified according to whether the writer was clearly responsible

for the situation, or whether the responsibility was ambiguous, following the

example of Ferguson et al. (2000). A situation was classified as ambiguous if

the person had no control over the situation or the person did not know how the

situation could be corrected. The categories for situation type and coping were

created based on the data, applying qualitative content analysis according to the

guidelines presented by Flick (2002). First the passages that were relevant for the

studied concept were identified, and then similar passages were bundled and

summarised further in order to reduce the data and reach a sufficient level of

abstraction. The analysis yielded three main categories: reparative behaviour,

chronic rumination and defences. The coding was made separately for each

category, according to whether the narrative included the reaction or not. In

addition, the narratives were grouped into four situation types: interpersonal

situations, achievement situations, norm violations and victim situations. A second

coder classified 30% of the narratives, and inter-rater agreement was between 91%

and 97% for all the categories. Differences were resolved by discussion.

Results

Based on the data, the guilt or shame situations described were divided into four

categories. Most narratives described interpersonal situations (62%): the respondents

Coping with guilt and shame 175

felt guilty or ashamed of not being a good friend, spouse, parent or relative. These

feelings could also be directed to strangers; there were several stories of guilt or shame

for not helping a drunken person lying in the street or not doing enough for poor

people. Another common theme was achievement or performance (13%): the person

felt guilty or ashamed for not studying or working hard enough, eating or drinking too

much or not keeping his or her home clean. These behaviours made the person feel

inadequate or dissatisfied with him or her self, and the transgression was more against

self than others. The third situation type was norm violations (14%): such behaviours

did not directly hurt others, but they were against societal or religious norms, for

example shoplifting, cheating on an exam or having premarital sex. The fourth

situation type was labelled as victimization (11%): these stories described situations in

which the respondent was a victim of emotional, physical or sexual abuse, or suffered

from some kind of psychological or physical illness, e.g. panic disorder. In these stories

the respondents were in a victim’s role, but still felt shame or even guilt over the

situation.

Combined guilt and shame occurred in 41% of the narratives: There were

negative self-evaluations and concern for how others perceive the self, but also

concern for others and for the consequences of one’s behaviour. In situations

involving pure shame (24%) the focus was on others’ thoughts and reactions and on

negative self-evaluations; however, in situations involving pure guilt (33%) these

were not mentioned, but the focus was on one’s responsibility for the behaviour’s

consequences. In most stories the respondent’s responsibility was clear (73%): the

writer knew how he or she should have behaved, and there was an obvious means by

which the writer could have affected the situation. In 27% of the narratives the

respondent’s responsibility for the situation was defined as ambiguous. For example,

in victim situations it is not reasonable to hold the victim responsible, even though

s/he may blame her or himself.

Coping was coded according to three categories: reparative behaviour, chronic

rumination and defences. Reparative behaviour included behaviour that was focused

on correcting the cause of the guilt or shame. This behaviour was different

depending on the situation. For example, if a parent felt guilty for spanking his or her

children, the person could try to correct the situation by apologising to the children

and avoiding such behaviour in the future. Another example could be somebody

who felt ashamed of having an unhealthy diet, which could be corrected by changing

eating habits for the better. The reparative behaviour category included also

prosocial behaviour that did not correct the original cause of guilt or shame, but was

motivated by these emotions; for example that a person gets too much change in a

café by mistake and does not return the money, but later gives money to charity to

alleviate guilt. Also intention to behave differently in similar situations in the future

were defined as reparative behaviour in cases where there was no means to correct

the situation, for example that the person one has not treated well has already passed

away. Chronic rumination was coded when the writer described suffering

considerably, and for a long period of time, from guilt or shame or indicated that

he or she had not yet been able to alleviate the emotion and still continued to suffer

176 M. Silfver

from it. Defences included externalising responsibility, minimising the importance of

the event, or avoiding certain thoughts, persons or situations. Reparation occurred in

52% of all situations, defences in 48% and chronic rumination in 30% of the

situations.

In order to get an overview of the data, optimal scaling was performed by SPSS.

This analysis creates a distance matrix based on similarities and dissimilarities

between objects. It is similar to multidimensional scaling, but it uses chi-square

metrics as a basis for the distance matrix, and therefore it can be applied to

categorical data. According to the created two-dimensional configuration (Figure 1),

ambiguous situations were close to chronic rumination and combined guilt and

shame, whereas unambiguous situations were close to guilt and absence of chronic

rumination. This dimension explained 34% of the variation. On the other

dimension, explaining 25% of the variation, shame was close to defences and

absence of reparation, whereas guilt and combined guilt and shame were nearly

equally close to reparation and absence of defences. The statistical significance of the

specific connections was tested by log-linear analysis. The interactions between

responsibility, emotion and coping were not significant for any of the three types of

coping (reparation, x 2

(2, N5112) 4,25, p5ns., chronic rumination, x 2

(2, N5112)

1,40, p5ns., defences, x 2

(2, N5112) 1,75, p5ns.). Against the expectations

derived from Tangney and Dearing (2002), chronic rumination, x 2

(2, N5112)

3,64, p5ns., and defences, x 2

(2, N5112) 1,52, p5ns., were not significantly related

to emotion. However, reparative behaviour was less likely in shame situations than in

guilt situations or in situations of combined guilt and shame, x 2

(2, N5112) 9,18,

Figure 1. Optimal scaling of coping with guilt or shame

Coping with guilt and shame 177

p50.01 (Table I). The ambiguity of responsibility had no effect on the likelihood of

defences, x 2

(1, N5115) 1,49, p5ns., or reparative behaviour, x 2

(1, N5115) 1,83,

p5ns., but ambiguous responsibility appeared to increase the likelihood of chronic

rumination, x 2

(1, N5115) 6,60, p50.01 (Table 2), in accordance with the view of

Ferguson et al.(2000). A more detailed examination of the context revealed that

chronic rumination occurred most often in victim situations (62%), and it was also

common in interpersonal situations (34%), but it was rare in norm violation (12%)

or achievement situations (13%). The three types of coping could all exist at the

same time, but the likelihood of reparation was smaller when defensive thinking was

used, x 2

(1, N5115) 20,08, p,.001. However, neither defences, x 2

(1, N5115)

0,12, p5ns., nor reparation, x 2

(1, N5115) 0,94, p5ns., were related to chronic

rumination. In addition, emotion and responsibility were related; in ambiguous

situations there was less shame-free guilt and more shame and combined guilt and

shame than in unambiguous situations, x 2

(2, N5117) 6,22, p5.045.

Discussion

The general structure of the data was very similar to that in the findings of Fontaine

et al. (2006), even though a different data collection method was used, and the

studied items differed somewhat. Guilt and shame were differentiated on one

dimension and interpersonal and intrapersonal orientations on the other dimension.

Both situational context and type of emotion were found to relate to coping with

guilt or shame. Shame was less likely to motivate reparative or prosocial behaviour

than guilt or a combination of guilt and shame. However, shame-free guilt did not

differ much from combined guilt and shame in terms of the likelihood of reparative

Table 1. Reparative behaviour according to emotion

Emotion

Guilt Shame Guilt and shame

Reparative or prosocial

behaviour as a coping

method

Yes 25 (66%) 8 (30%) 28 (60%)

No 13 (32%) 19 (70%) 19 (40%)

Total 38 27 47

Table 2. Chronic rumination according to responsibility

Responsibility

Unambiguous Ambiguous

Reacting by chronic

rumination

Yes 21 (25%) 15 (50%)

No 64 (75%) 15 (50%)

Total 85 30

178 M. Silfver

or prosocial behaviour. This conflicts with Tangney’s view of shame being a

corruptive element in a guilt experience. In some cases, shame combined with guilt

appeared to produce more of a profound preoccupation with one’s motives or

personality characteristics than would mere guilt, as Lindsay-Hartz et al. (1995)

suggested. For example, a 28-year-old woman wrote about not helping a tourist to

buy a ticket in a tram:

… I felt guilty for not doing that (helping) … I’m annoyed by indifferent people who

don’t help, who pretend not to notice when somebody has problems. Now I felt that I

had become one of them, and I felt ashamed of that feeling. I decided that, from now

on, I would keep my eyes open and help every time I can …

In most stories the unique contribution of shame to the outcome was not explicated

so clearly, so it is impossible to say if feeling both guilt and shame for transgression

or negligence would have some kind of advantages compared to guilt alone.

However, it was clear that ‘pure guilt’ was not more likely to motivate reparation

than a combination of guilt and shame. In addition, there were cases in which fear of

shame was said to motivate reparative behaviour. Shame is a painful emotion, and

the willingness to correct one’s shame-causing behaviour in order to avoid shame in

the future seems reasonable. Still, it must be acknowledged that if the change in

behaviour is motivated only by fear of how others perceive the self and does not

reflect internalized morality, the change may be more superficial and temporary than

a guilt-motivated change. Reparative behaviour was the only reaction type that

differed significantly according to emotion. Against Tangney’s (1995) theory,

chronic rumination was found to occur regardless of the emotion, and reparative

behaviour was found to be unrelated to chronic rumination: in some cases reparative

behaviour was not enough to alleviate guilt or shame, but the respondent felt that

s/he should do more in order to alleviate the emotion.

In addition to the type of emotional reaction, context also was found to be

relevant. Chronic rumination was more likely in situations in which the writer’s

responsibility for the situation was ambiguous, according to the view of Ferguson

et al. (2000). A more detailed analysis of the context revealed that chronic

rumination occurred most frequently in victim situations and also often in

interpersonal situations, but it was rare in norm violation or achievement situations.

However, Hoffman (2000) seems to be right in his claim that all forms of guilt

motivate prosocial behaviour: reparative behaviour did not depend on the type of

situation. The relation between prosocial motivation and psychological well-being is

not completely straight-forward; usually considering others can be seen as positive,

but an excessive sense of responsibility can have psychological costs. A good example

of this was a narrative by a 27-year-old woman who wrote about having been raped

as a teenager:

… I carried on my life feeling guilty and ashamed of almost everything around me. I

started to sacrifice myself for others and live my life though others … I felt bad to the

core and I didn’t understand the importance of my own needs, desires and hopes.

To be adaptive, the feeling of responsibility has to be reasonable in the given

situation (Ferguson et al., 2000).

Coping with guilt and shame 179

The use of defences was not found to depend on context or emotional experience.

However, reparative behaviour was less likely when defensive thinking was used. Still

defences cannot always be labelled as maladaptive. If the person is clearly

responsible for the situation, blaming others or trivialising the situation can be seen

immoral, but the same defences may be essential in terms of mental health for

someone blaming her or himself unreasonably for things s/he cannot change.

There are also some limitations of this research which need to be addressed. First,

most respondents were women, and therefore the results describe mainly women’s

experiences. It is possible that the findings would have been different if the sample

had comprised more men. However, analysing one’s emotions is considered

feminine in Finnish culture, and obtaining more narrative data on men’s guilt and

shame experiences could be difficult, although interesting. In addition, the data is

not representative of the Finnish population generally, because the respondents were

attending university courses, and they were willing to share their experiences without

any compensation. According to Baumeister, Stillwell and Wotman (1990), the

study of autobiographical narratives has both strengths and weaknesses as a

psychological methodology. It is possible that the respondents fail to report their

thoughts, emotions and behaviour accurately or may even deliberately lie. However,

narratives can provide insight into how people construct their experiences and what

kind of motivations they have. Furthermore, the advantage of narrative data is that

the situations are authentic; it would be unethical to include serious and

traumatising incidents in a scenario-based measure or in laboratory research.

In spite of these limitations, the data provide interesting examples of different

possible patterns of emotion and behaviour. What are the implications of these

observations for moral education? First, the narratives show that guilt and shame

play an important role in moral behaviour, and experiencing and coping with these

emotions should be within the focus of moral education. Different aspects of guilt

and shame experiences may cause problems in terms of moral behaviour. Some

people do not feel bad enough to correct their immoral behaviour, even though they

realise that what they do is morally wrong. Others may be overwhelmed by the sense

of responsibility, or they may be too distressed to behave in a constructive way.

According to the present study, shame without guilt elements appeared to diminish

the likelihood of moral or prosocial behaviour. Guilt-free shame can be seen to

reflect a self-centred orientation, because the person’s main concern is how the

situation affects him or herself. Feeling guilty requires internalised morality that

forces the person to rise above self-centred concerns. The ability to feel guilty is

usually seen to develop during childhood. Understanding and appreciating others’

emotions and perspectives, and the broader implications of one’s behaviour, can be

enhanced by parenting practices that highlight others’ perspectives in disciplinary

encounters (Hoffman, 2000). According to Hoffman (2000), guilt arises when the

child feels empathy towards the other person and understands that she or he is

responsible for the other’s distress. Therefore explaining to the child why certain

behaviour is not acceptable is essential. However, there is no evidence that direct

shaming or denigrating the child would be necessary or beneficial when disciplining

180 M. Silfver

the child. In addition to disciplinary methods, the example provided by parents or

other educators is important. By observing others, children learn ways to feel and

react in different situations and ways to express emotions (Tangney & Dearing,

2002). Furthermore, sometimes the child may be motivated to reparation, but does

not know how to do it. In this type of situation teaching practical social skills can be

helpful. It has also been found to be important that a child learns to regulate his or

her emotional reactions: if the emotional arousal is too uncomfortable, it is difficult

to concentrate on others’ perspectives (e.g. Eisenberg, 2000). Parents who express

emotions to their children and acknowledge and name the children’s emotions help

them to learn to regulate emotions. This has been found to be important for the

development of sympathy and prosocial tendencies (Eisenberg, 2000). For example,

acknowledging one’s shame can prevent defensive reactions like anger (Ahmed &

Braithwaite, 2004). Eisenberg, Wentzel and Harris (1998) emphasise that

interventions are most effective if they recognize individual differences in

emotionality and regulation. Interventions that promote understanding of others’

emotions may be most beneficial to unemotional children, who tend to misinterpret,

ignore or distort others’ emotions. However, for children who are prone to intense,

negative emotions, enhancing emotional responsivity is not useful. They probably

benefit most from programs concentrated on techniques for regulating emotions,

which help them to cope with their emotions and prevent over-arousal, which in turn

is likely to increase other-orientation and prosocial behaviour.

The finding that guilt-free shame is less likely to motivate reparative behaviour

does not mean that all shame experiences should be avoided. Sometimes shame-free

guilt may not be appropriate, for example when the transgression is serious. Gestures

that demonstrate shame may have a positive effect on others; they see that the person

takes the transgression seriously. A healthy self-esteem is important for adaptive

social functioning, but it does not mean that a person never evaluates his or her

personality critically. It is important to be aware of the strengths and weaknesses in

one’s personality, and strive for improving oneself as a person, even though changing

personality is more complicated than changing specific behaviour. Educators can

help children to learn to evaluate themselves realistically without being overly self-

critical, and provide means for positive change.

In the light of the findings of this study, it is also essential to understand the limits

of one’s responsibility. Children especially need to be clearly told what is expected

from them and what their responsibilities are. Children tend to think egocentrically

and have difficulties understanding causal relations, and consequently have a

tendency to make false attributions of their blameworthiness (Hoffman, 2000). For

example, children may think that they are responsible for their parents’ divorce,

unless they are convinced that it is beyond their control and they are not to blame.

Even adults have difficulties finding appropriate limits for their responsibility:

Gilligan (1982) suggested that, due to the feminine gender role emphasising caring

and nurturing, women sometimes take responsibility for others at the expense of

their own needs, to the extent that it becomes psychologically harmful. Examples of

this were found in the narratives of the present study.

Coping with guilt and shame 181

In future research it could be fruitful to concentrate on coping with guilt or shame in

different contexts. This research might examine the antecedents of effective and

ethical coping and how personality characteristics and situational factors interact to

facilitate or inhibit coping. It could be assumed that coping with guilt and shame

depends to a certain degree on a person’s dispositions and skills: is the person able to

regulate her or his emotions and does she or he have the necessary interpersonal skills

to solve the situation? At the same time, there is no doubt that the social environment

plays an important role in coping; how others behave towards the person, whether or

not they induce guilt or shame, and in what kind of situations they do it?

Understanding the interplay between individual behaviour and the situational

dynamics of guilt and shame remains a challenge for psychological research.

References

Ahmed, E. & Braithwaite, V. (2004) ‘What, me ashamed?’ Shame management and school

bullying, Journal of Research in Crime and Delinquency, 41(3), 269–294.

Bagozzi, R. P., Verbeke, W. & Gavino, J. C. Jr. (2003) Culture moderates the self-regulation of

shame and its effects on performance: the case of salespersons in the Netherlands and the

Philippines, Journal of Applied Psychology, 88(2), 219–233.

Barrett, K. C. (1995) A functionalist approach to shame and guilt, in: J. P. Tangney

& K. W. Fischer (Eds) Self-conscious emotions: the psychology of shame, guilt, embarrassment

and pride (New York, Guilford Publications), 64–113.

Baumeister, R. F., Stillwell, A. M. & Heatherton, T. F. (1995) Personal narratives about guilt: role

in action control and interpersonal relationships, Basic and Applied Social Psychology,

17(1&2), 173–198.

Baumeister, R. F., Stillwell, A. & Wotman, S. R. (1990) Victim and perpetrator accounts of

interpersonal conflict: autobiographical narratives about anger, Journal of Personality and

Social Psychology, 59(5), 994–1005.

Bybee, J., Merisca, R. & Velasco, R. (1998) The development of reactions to guilt-producing

events, in: J. Bybee (Ed.) Guilt and children (San Diego, CA, Academic Press), 185–209.

Bybee, J. & Quiles, Z. N. (1998) Guilt and mental health, in: J. Bybee (Ed.) Guilt and children (San

Diego, CA, Academic Press), 269–291.

Eisenberg, N. (2000) Emotion, regulation and moral development, Annual Review of Psychology,

51(1), 665–697.

Eisenberg, N., Wentzel, N. M. & Harris, J. D. (1998) The role of emotionality and regulation in

empathy-related responding, School Psychology Review, 27(4), 506–521.

Ferguson, T. J. & Crowley, S. L. (1997) Measure for measure: a multitrait-multimethod analysis

of guilt and shame, Journal of Personality Assessment, 69(2), 425–441.

Ferguson, T. J. & Stegge, H. (1998) Measuring guilt in children: a rose by any other name still has

its thorns, in: J. Bybee (Ed.) Guilt and children (San Diego, CA, Academic Press), 19–67.

Ferguson, T. J., Stegge, H., Eyre, H. L., Vollmer, R. & Ashbaker, M. (2000) Context effects and

the (mal)adaptive nature of guilt and shame in children, Genetic, Social & General Psychology

Monographs, 126(3), 319–348.

Fischer, A. H., Manstead, A. S. R. & Mosquera, P. M. R. (1999) The role of honour-related

versus individualistic values in conceptualizing pride, shame and anger: Spanish and Dutch

cultural prototypes, Cognition and emotion, 13(2), 149–179.

Flick, U. (2002) An introduction to qualitative research (London, Sage).

Fontaine, J. R. J., Luyten, P., De Boeck, P. & Corveleyn, J. (2001) The test of self-conscious

affect: internal structure, differential scales and relationships with long-term affects,

European Journal of Personality, 15, 449–464.

182 M. Silfver

Fontaine, J. R. J., Luyten, P., De Boeck, P., Corveleyn, J., Fernandez, M., Herrera, D., Ittzés, A.

& Tomcsányi, T. (2006) Untying the Gordian knot of guilt and shame: the structure of guilt

and shame reactions based on situation and person variation in Belgium, Hungary and Peru,

Journal of Cross-Cultural Psychology, 37(3), 273–292.

Gilligan, C. (1982) In a different voice: psychological theory and women’s development (Cambridge,

MA, Harvard University Press).

Harris, N. (2003) Reassessing the dimensionality of the moral emotions, British Journal of

Psychology, 94(4), 457–473.

Hoffman, M. L. (2000) Empathy and moral development—implications for caring and justice (New

York, Cambridge University Press).

Izard, C. E. (1971) The face of emotion (New York, Appleton-Century-Crofts).

Kitayama, S., Markus, H. R. & Matsumoto, H. (1995) Culture, self and emotion: a cultural

perspective on self-conscious emotions, in: J. P. Tangney & K. W. Fischer (Eds) Self-

conscious emotions: the psychology of shame, guilt, embarrassment and pride (New York, Guilford

Publications), 439–464.

Lindsay-Hartz, J., de Rivera, J. & Mascolo, M. F. (1995) Differentiating guilt and shame and their

effects on motivation, in: J. P. Tangney & K. W. Fischer (Eds) Self-conscious emotions: the

psychology of shame, guilt, embarrassment and pride (New York, Guilford Publications),

274–300.

Lutwak, N., Panish, J. B., Ferrari, J. R. & Razzino, B. E. (2001) Shame and guilt and their

relationship to positive expectations and anger expressiveness, Adolescence, 36(144),

641–654.

Luyten, P., Fontaine, J. R. J. & Corveleyn, J. (2002) Does the Test of Self-Conscious Affect

(TOSCA) measure maladaptive aspects of guilt and adaptive aspects of shame? An empirical

investigation, Personality and Individual Differences, 33(8), 1373–1387.

Mason, J. (2002) Qualitative researching (London, Sage).

Mosquera, P. M. R., Manstead, A. S. R. & Fischer, A. H. (2000) The role of honor-related values

in the elicitation, experience and communication of pride, shame and anger: Spain and the

Netherlands compared, Personality and Social Psychology Bulletin, 26(7), 833–844.

Quiles, Z. N. & Bybee, J. (1997) Chronic and predispositional guilt: relation to mental health,

prosocial behavior and religiosity, Journal of Personality Assessment, 69(1), 104–126.

Tangney, J. P. (1995) Guilt and shame in interpersonal relationships, in: J. P. Tangney

& K. W. Fischer (Eds) Self-conscious emotions: the psychology of shame, guilt, embarrassment

and pride (New York, Guilford Publications), 114–139.

Tangney, J. P. & Dearing, R. L. (2002) Shame and guilt (New York, Guilford Publications).

Wallbott, H. G. & Scherer, K. R. (1995) Cultural determinants in experiencing shame and guilt,

in: J. P. Tangney & K. W. Fischer (Eds) Self-conscious emotions: the psychology of shame, guilt,

embarrassment and pride (New York, Guilford Publications), 465–487.

Coping with guilt and shame 183