BritishJournal of Medical Psychology (2001), 74, 1–21 Printedin GreatBritain 1 q 2001The British Psychological Society

Shameandnon-d isclosure:Astudyofthe emotionalisolationof peopler eferredfor psychotherapy

James Macdonald* SalomonsCentre for Applied Socialand PsychologicalDe velopment, Kent, UK

Ian Morley Departmentof Psychology, University of Warwick, UK

Thirty-fourpeople referred to an NHS psychotherapydepartment were given a modiŽed formof Oatley and Duncan’ s (1992)emotion diary which included questions about whethereach recorded emotion had been subsequently disclosed to anyone (for example apartner,friend or professional). One week later the diaries were collected and participantsinterviewed. Interviews focused, among other things, on reasons for non- disclosureof recorded emotional experiences and the relationship between shame and non-disclosure.The results indicated that a majorityof the emotional incidents recordedin the diaries were not disclosed (68%). This result contrasts with studies onnon-clinical samples in which only approximately 10% of everyday emotions are keptsecret. Qualitative analysis of the interview data revealed that participants appearedto behabitual non-disclosers of emotional and personal experiences and that non-disclosurewas related to the anticipation of negative interpersonal responses to disclosure(in particular labelling and judging responses) in addition to more self- criticalfactors including shame. It issuggested that these results add to the existing literatureon shame by illustrating the interpersonal effects of shame in a clinical sample.

There are many unresolved issues regardingthe deŽnition of shame. Ina recent review of theories Gilbert (1998)argues that adeŽnition whichcaptures shame most closely is that it is ‘an inner experience ofself as an unattractive social agent,under pressure to limit possible damage to self via escape orappeasement’(p. 22). Shame can bedifferentiated from guilt where the action tendency promotes reparation andthe focus ofattention is outside the self (Lewis, 1971;T angney,Miller,Flicker, &Barlow,1996).Shame can be associated either with aconsciousness ofhowone is seen byothers (Sartre, 1943)or with a negative self-evaluation, andthese two different facets ofshame have been termed external shame and internalizedshame respectively (cf. Gilbert, 1998).While in practice these two cognitive domains are likely to behighlycorrelated (Lewis, 1971)this is notnecessarily the case (Gilbert, 1998).Perhaps the least controversial characteristic of shame is that it is

*Requestsfor reprintsshould be addressedto James Macdonald,Salomons Centrefor AppliedSocial andPsychological Development, David Salomons Estate,Broomhill Road, Southborough, T unbridgeWells, Kent, TN3 0TG, UK. 2 JamesMacdonald andIan Morley associated with adesire to hide or conceal the self oraspects of the self (Barrett, 1995; Lindsay-Hartz, 1984;Mollon, 1984). Inrecent years studies have linked atendency toexperience shame with the incidence ofpsychopathology(Andrews, 1995, 1997; Gilbert, Pehl, &Allan, 1994;Harder, 1995; Sanftner, Barlow,Marschall, &Tangney,1995;T angney,Burggraf,& Wagner,1995; Tangney,Wagner,&Gramzow,1992);however Alexander, Brewin,V earnals, Wolff, and Leff (1999)found to the contrary that pronenessto guilt rather thanshame was associated with depression.Most of this research has been cross-sectional andquestionnaire-based andto date there appears tohave been noattempt to examine the role shame plays in the interpersonal lives of psychologically distressed people. Anaspect of psychologically distressed people’s interpersonal lives that seems especially likely tobeaffected bythe experience of shame, andthe accompanying impulse of hidingor concealment, is that of emotional disclosure. Bothcommon sense andempirical evidence suggestthat emotional disclosure is an important componentof clinical practice. Stiles (1995)has reviewed anumberof studies which suggestthat all types of psychotherapyare characterized byhigh levels of personal disclosure byclients andthat personal disclosure bythe client is in turncorrelated with measures of goodpsychotherapy process. Aconsumer studyby the Mental Health Foundation(1997) indicated that aprimary need of mental-health patients whenin distress is to have ‘someone to talk to’. Thisis consistent with the protective effect of havinga close conŽding relationship in the presence of other factors conferring vulnerability to psychopathology(Brown & Harris, 1978).Pennebaker andcolleagues (e.g. Pennebaker,1993) have demonstrated that the disclosure of traumatic experiences appears to have physical-health beneŽts (includingdecreases in visits tophysicians andan enhancement ofimmune function) whendisclosure includes the emotional response to the experience. There is already some evidence that shame may beassociated with emotional secrecy in non-clinical populations.Research byRime ´,Finkenauerand colleagues has revealed that disclosure of everyday emotional experiences occurs approximately 90%of the time (Rime´,Mesquita,Philippot, & Boca, 1991;Rime ´ et al.,1994).Follow-up research designedto identify characteristics of the 10%of emotions that were notdisclosed suggestedthat ‘secret’emotions are associated with the desire to avoid shame andother unpleasant social emotions suchas guilt andembarrassment (Finkenauer &Rime´, 1996; Finkenauer,Rime ´,&Lerot,1996). Finkenauer and her colleagues have notfocused onthe role ofshame per se in the non-disclosure of emotional experiences. Nor didthey examine the disclosure decisions of‘normally occurring’emotions. Finally,as notedabove, their workwas conductedin a non-clinical sample. Thecurrent studyaims to extend the literature onshame andpsychopathology and shame anddisclosure byexamining the impact of shame onan aspect of the daily interpersonal lives of psychologically distressed adults—namely their non-disclosure of speciŽc negative emotional experiences. Given the lack of research in this area, it was decided that aspeciŽc focus onthe role of shame shouldbe accompanied bya more open-endedinquiry into factors associated with non-disclosure.This, it was hoped, wouldresult in aholistic andcontext-sensitive view of the role ofshame in emotional non-disclosure. Thedesire to develop an inclusive andcontext-sensitive understandingof participants’ Shameand non-disclosure 3 decisions notto disclose emotions led to the adoptionof aqualitative research method based onthe ‘grounded’techniques described byPidgeon and Henwood (1996). This procedure enables the researcher to develop atheoretical understandingwhich is based closely onparticipants’ own accounts andwhich can encompass unforeseen factors which mightotherwise beobscuredby the researchers’ a priori constructs.

Method

Materials AmodiŽed form of Oatleyand Duncan’ s (1992)emotion diary was used to obtain accounts of shame,guilt, hatredand disgust. This method is based on the assumption that, while not all affective experiences are accessibleto introspection, people are able to identify accurately many of their more salient everyday emotionalexperiences (see Oatley and Duncan (1992) for further discussion of thisissue) including shame. A reviewby Macdonald (1998) suggests that, in spite of the fact that few instruments intended to measure shameactually ask participants directly about their experiences of shame (Andrews, 1998), the available evidencesuggests that both clinical and non-clinical samples will voluntarily report experiences of shamein a conŽdential research setting. Oatley and Duncan’ s (1992)development of emotion diaries represented an attemptto obtain more reliable accounts of subjectiveemotional experiences by minimizing retrospective biases,which would seem more likely in studies where people are simply asked to recall an instance of the emotionof interest,as forexamplein Wicker,Payne, and Morgan (1983). Oatley and Duncan (1992) provide someevidence for the validity of thismethod in a studycomparing participants’ diary record of theirown emotionswith an independent record of their emotions provided by their partners. This demonstrated an 84%agreement as totheoccurrence of incidents. Thediary was structured and asked participants to recordspeciŽ c detailsabout each emotion as soonas possibleafter the emotion had been experienced. Participants were asked to Ž llin the diary questions for the Žrstinstance of anyof thefour emotions of shame,guilt, hatred and disgust that they experienced in the7 daysafter they had been given the diary. The diary included a numberof questionsabout the disclosure or non-disclosureof each emotion that had been recorded. Of relevanceto the current report was the question ‘Didyou tell anyone about this?’ and two follow-up questions asked if thereply was negative ‘ Wereyou afraid abouthow others might see you if you told them?’ and ‘ Didthe thought of tellinganybody make you feelany shame?’Full details and Ž ndingsfrom the diary are reported in Macdonald (1999). Afollow-upsemi-structured interview was carried out in which participants were asked to provide informationabout (1) the reasons for the non-disclosure or disclosureof eachemotion they recorded in the diary,(2) shame and related feelings associated with anticipated disclosure of eachemotion, and (3) theeffects ofdisclosure or non-disclosure of each emotion on their relationships with other people. The interview protocolfor non-disclosed emotions is included in Appendix1. Thesequestions were designed as afollow-up tothe ‘ yes/no’questions about non-disclosure included in the diary. The intention was to obtain a richer accountof factorsassociated with the decision to keep the emotional experience private, the role of shamein thisprocess and the impact of non-disclosure on theparticipants’ relationships. In thispaper, due to space limitations,only the Ž ndingsrelating to (1) and (2) (reasons for non-disclosure, including shame) will be reported.It was hoped that by asking these questions for eachemotion recorded in thediary the information providedwould be morereliable and more valid as aresultof beinggrounded in concrete instances.

Participants Participantswere people referred to an NHS psychotherapyout-patient clinic which specialized in the provisionof psychodynamicallyinformed psychotherapy .Themajority of referralsto the service were made bylocalGPs and38% (48/ 126)of the people invited to take part completed the diary part of thestudy. Of the62% (78/ 126)who did not complete it, 54% (42/ 78)said they did not wish to takepart, 36% (28/ 78) failedto turnup to one or other of theappointments (most frequently the second) and the remaining 10% (8/ 78)of non-completershad other reasons for not attending one of theappointments. Of thosewho did complete the study, 20.8% (10/ 48)reported that they had not experienced any of the 4 JamesMacdonald andIan Morley targetemotions. Four of the remaining 38 participants failed to provide an audiotaped interview (one participantdid not want her interview taped, two participants were unable to attend the interview appointment,and in onecase, the tape-recorder did not work). The Ž nalsample therefore consisted of 34 psychotherapyreferrals, which was 27% (34/ 126)of those invited to take part. Of these34 participants, 73.5%(25/ 34)werewomen. 55.9% of thosewho provided interviews (19/ 34)had,according to their referral letters,medical notes or what they said in the interview ,sufferedchildhood sexual abuse (CSA). Studies have foundthat between 26 and40% ofpsychiatricin- andout-patients have suffered CSA (Drauker,1992); 60% (15/25)of thewomen in the Ž nalsample had suffered CSA and44.4% (4/ 9)ofthemen. The average age of theparticipants was 37.2 years. Itis not clear exactly why there was such a lowparticipation rate; however, the demanding nature of the research(monitoring emotional experiences for a weekand coming in to the clinic especially for the interview),the requirement to focuson painfuland distressing emotional experiences, anxiety associated with beingassessed for psychotherapy and possible feelings of shame associated with disclosure to theresearcher, mayall have played a role.The low level of participationraises questions about the representativeness of the Žnalsample; 70% (63/ 90) 1 ofthose invited to take part who either did not complete the study, did not experienceone of the target emotions or did not provide an interview for other reasons, were women comparedwith 73.5% in the Ž nalsample, suggesting that there was no gender differential drop-out rate. About36% (31/ 86) 2 ofthosewho did not completethe study ,didnot experience one ofthetarget emotions ordidnot provide an interviewfor other reasons, were identiŽ ed as having suffered CSA intheir medical notesor referralsmade to theclinic. This contrasts with 55.9% of thosewho provided interviews, although a numberof thelatter only identiŽ ed themselvesas havingsuffered CSA inthe interview itself. However, only 22% (2/9)3 ofthose who did not provide an interview because they reported not experiencing any of the targetemotions in thediary were identiŽ ed as having experienced CSA. Thissuggests that people who had sufferedCSA arelikely to havebeen somewhat over-represented in theŽ nalsample.

Procedure Participantswere Ž rstcontacted by letterin which they were given a patientinformation sheet explaining the studyand asked to returna formsaying whether they wished to participate. When permission was received therewas usually a 2-weekperiod prior to their Ž rstmeeting with the researcher. They were seen initially for approximately15 minutes when the diary was explained and an appointment made 1 weeklater for the returnof thediary and a follow-upinterview .Forall but three of the participants, this Ž rstappointment took placeimmediately prior to their assessment appointment for psychotherapy 4.Thereason for this was that the studyaimed to examine processes of disclosure and non-disclosure before these had been substantially inuenced by treatment.However, as notedabove, proximity with the stressof the psychotherapy assessment mayhave contributed to lowparticipation rates. Whenparticipants returned to theclinic for the follow-up interview ,theinterviewer went through the responseswritten in the diary with the participant, which functioned partly as a meansof clarifying the writtenresponses and partly as a meansof bringing the recorded emotion back into the participants’ awareness.The interviewer then went through the interview questions for each emotion that was recorded in the diary. Thestudy was approved by thelocal NHS ethicscommittee and all participants signed a consentform whichstressed that participation was voluntary. After the interview, participants were thanked and the backgroundand purpose of thestudy were explained to them.Following analysis of theresults,participants andclinicians in theservice were sent a summaryof the main Ž ndings.

Analytic strategyforthe interviewdata Inthispaper only analysis of datarelating to reasons for non-disclosure of emotionsrecorded in the diaries is reported.The analysis was conducted inductively by developing codes in a ‘grounded’manner from the

1 Thegender of two peoplewho were invitedbut did not take part in thestudy was notestablished. 2 Notes couldnot be traced for fourpeople invited who did not take part in the study . 3 Notes couldnot be traced for one of theseindividuals. 4 Threeparticipants were ina groupthat started before Isaw them. Shameand non-disclosure 5 descriptionsprovided by participants (Pidgeon & Henwood,1996). This procedure enables the researcher to developtheoretical ideas which are based closely on participants’own accounts and should allow factors that areimportant in the participants’ social world to emerge in the developing theory. It seems particularly appropriateto use such a methodin thiscase because there appear to befewalternative explanations offered forwhy participants fail to revealemotional experiences. Interviewswere transcribed by theŽ rstauthor, and all information relating to reasons for non-disclosure wasmarked. This material was extracted from the interviews and labelled with the number of theparticipant itcame from. Using the material on reasons for non-disclosure, an open-ended coding system was then developedin which categories were generated to describe reasons given for non-disclosure. These categories werereŽ ned until a categorywas developed for all recurring reasons for non-disclosure mentioned by the participants.A descriptionof each category was written and a notewas made of theway it appearedto be linkedto othercategories. Where possible these lower-level categories were organized into broader and more abstractcategories. An example of coding from one interview is includedin Appendix2. Anumberof techniques were used to enhance the validity of the conclusions drawn from the interview data.First, examples from the interviewsare presented so that readers can see for themselves the relationship betweencategories and the source data, as recommended by Elliott, Fischer, and Rennie (1999). In the originalanalysis this process was taken a stepfurther by using data-display matrices as a wayof illustrating thecategories in a systematicway (Miles & Huberman,1994). The data-display matrix for the category of ‘habitualnon-disclosure’ is includedin Appendix3; however, other display matrices have not beenincluded heredue to spacelimitations. They are presented in Macdonald(1999). A secondapproach has been to state howrepresentative each category is of thesample as awholeby notingthe proportion of participantswho madecomments which were coded in each category. Third, the original analysis included a negativecase analysis(Y in,1989). In fact, only two participants had reasons for non-disclosure which diverged markedly fromthe sample as awhole.Discussion of these cases has not been included here due tospace limitations but canbe foundin Macdonald (1999).

Results

Results fromthe diaryre lating ton on-disclosureo femotions Ofthe emotions reported inthe diaries, 68%(51/ 75)were notdisclosed. Thiscontrasts with the discovery ofRime´ et al.(1991,1994) that between 4%and10% of emotions recorded bya variety ofnon-clinical populationswere notdisclosed to others.(Rime ´ and his colleagues studiedall the emotions in the current diary,with the exception of hatred, whichwas the emotion whichwas most disclosed inthe current study.) Whenparticipants didnot disclose an emotion, the diary asked ‘didthe thoughtof telling anybodymake youfeel any shame?’; 90.9%(10/ 11)of non-disclosed instances of shame were given a‘yes’rating for this question.However theproportionwas quite high for the other three emotions as well—inall, 65%(26/ 40)of the other non-disclosed emotions were associated with shame at the thoughtof telling anyone (66.7%(12/ 18) for guilt,72.7% (8/ 11)for hatred and54.5% (6/ 11)for disgust).This appears to supportthe notionthat in this clinical populationexperiences of shame are associated with non-disclosure ofemotional experiences. Whenparticipants hadnot disclosed an emotion they were also asked,‘ were youafraid abouthow others mightsee youif youtold them?’Overall 63.2%(24/ 38)of the undisclosed emotions were associated with positive responses to this question(50% (8/16)for guilt,72.7% (8/ 11)for hatred,72.7% (8/ 11)for disgustand 72.7% (8/ 11)for shame). Thissuggests that, in keepingwith feelings of shame, non-disclosure is associated with theperception that other people will regard the individual less favourably if they disclose the emotional experience. 6 JamesMacdonald andIan Morley Results fromthe interviews relatingtore asonsfornon-disclosure Thispart of the analysis uses material from the 27interviews in whichparticipants spokeabout an instance in which they didnot disclose an emotion to anyone else (79.4%,27/ 34,of the participants in the studydiscussed at least one emotion which they hadnot disclosed). Thematerial comes mainly from participants’responses to the questions,‘ Ifyoucan, can youexplain whyyou chose notto tell anyone?’and ‘ Doyou thinkthat if youtold somebodyyou would feel stupid,or silly,or ashamed of yourself?’. Categories developed in a‘groundedmanner’ from the interviews will be presented along with atable summarizing whichcodes were given to each participant. Because of the comparatively large numberof categories whichemerged in the analysis, alist of the categories relating to reasons for non-disclosure is included in Table 1.

Table 1.Themain categories ofreasons fornon-disclosure of emotionalexperiences recorded in the diary Maincategory Sub-category Habitualnon-disclosure (81.%, 22/ 27) Anticipatedresponse to disclosure Negativerecipient responses (e.g.labelling, judging andblaming) (70.4%, 19/ 27) Fearof upsettingor burdeningothers (55.5%,15/ 27) Unhelpfulpositive responses ofothers (33.3%,9/ 27) Others notunderstanding (33.3%, 9/ 27) Lackof interest or attention(25.9%, 7/ 27) Disclosure pointless(25.9%, 7/ 27)

Shameand other factors associated withthe self Shameand self-conscious emotions (74.1%,20/ 27)—shame (18.5%,5/ 27)—guilt Out ofcharacter (40.7%,11/ 27) Inabilityto justifyfeelings and experiences (37%, 10/27) Ownresponsibility (33%, 9/ 17) Rejectionof ownfeelings (18.5%, 5/ 27) Reluctanceto experienceunpleasant emotions or memories (22.2%,6/ 27) Isolation(29.6%, 8/ 27)

Habitual non-disclosure .Responses to the interview question,‘ If youcan, can youexplain whyyou chose notto tell anyone?’included many comments that suggestedthat non- disclosure was habitual. Thiswas often described byparticipants as ageneral tendency presumably relating to awide range of emotional experiences: Shameand non-disclosure 7 ‘Inormallykeep things to myself (P5 5).

’I’mnotone for telling people how I’ mfeeling’(P33).

’Isomehowkeep it allbottled up’ (P18). Insome cases non-disclosure was described as apropertyof the kindof personthey were:

’I’maveryprivate person’ (P8). However, in other cases it seemed to be more speciŽc to aparticular experience or emotion, as inthe following extracts: ‘Youknow,to meit’ s somethingprivate and I justdon’ t wantto shareit withanybody’ (P22).

‘Idon’t Žnditeasy to share,that particular feeling’ (P15). Overall 81.5%(22/ 27)of the participants whodiscussed anundisclosed emotional experience indicated that non-disclosure was arecurrent or habitual pattern.This corroborates the pattern of non-disclosure foundin the diary part of the studyand it underlines the contrast between the studyparticipants andthe non-clinical populations studiedby Rime ´ andhis team who,as notedabove, foundthat disclosure is anormal characteristic of the aftermath of emotional experience. These Žndingssuggest that such ‘emotional isolation’might be asigniŽcant componentin these participants’emotional distress. Italso highlightsthe importance of an understandingof whythe participants chose notto disclose.

Anticipatedresponsesofothers .Manyof the reasons given for notdisclosing could beloosely groupedunder a broader category relating to participants’expectations of howother people wouldrespond if they disclosed.

Negative recipient responses(includinglab elling,judging and blaming) .Themost frequent of these anticipated responses were negative andthese generally involved labelling, judging orblaming.Participants feared that,if they disclosed, others wouldsee them as ‘barmy’, ‘crackers’, ‘stupid’, a‘freak’or simply ‘judge’them. The following comments capture the spirit of these responses: ‘toa normalperson in the street, you know ,Imean,the basic thing is whenyou turnedyour back they say‘ ‘He’s roundthe bend’ ’, youknow .So,uh, you tend not to, you bottle it allup, you know .Uh, basicallyfor your own self-respect I suppose.Y ouknow,togiveyou what little respect you’ ve gotleft’ (P38). ‘Icouldn’t seemyself telling anybody at all. Because, you see, because, people don’ t knowthe things, theycan’ t judgeme by them.So Iwouldn’t tellthem’ (P16). Afew said orimplied that they thoughttheir disclosure wouldlead to arguments.P38 said that if he spokeabout his experience tohis wife she mightnot agree with him,in which case heis ‘likely to goup the wall again’and this will ‘double’his problem.Some

5 Refers to ‘participant5’ in the study . 8 JamesMacdonald andIan Morley said that other people were likely totell them that their feelings were wrong,for instance P10who thought his wife wouldsay ‘Isupposethis is another bloodydepression’ , andP8 whothought her friends wouldtell her what she was feeling was wrong. Overall, 70.4%(19/ 27)of the participants invokeda fear of this kindof negative recipient response to explain whythey chose notto disclose. Thecommon denominator inall these anticipated responses seems to be that the recipient of the disclosure will respondin a clearly invalidating manner—generally evaluating the participant less favourably.

Fearo fupsettingorburdening others .Abouthalf (55.5%,15/ 27)of the participants mentioned notwanting to upsetor burdenothers bytelling them aboutthe experience:

‘Idon’t reallyfeel I oughtto burden people with my depression, or myproblems’ (P14).

‘Ijustfeel like I’ mjustputting on people and I’ mnot,you know ,they’ve gotbetter things to do, than listento me, wallowing in self-pity ,youknow’ (P15).

Unhelpfulp ositive responses .Inaddition to burdeningor damagingothers, a number (33.3%,9/ 27)of non-disclosingparticipants talked abouthow other people might respondin an apparently positive way whichwas nevertheless perceived as unhelpfulas the following example illustrates: ‘Ithinkthat a littlewhile ago, a fewyears ago, when I Žrstsaw [name of psychiatrist]and then sort of toldmy mum about it, my parents said ‘ ‘Don’t beridiculous, pull yourself together. Y ou’vegot everythinggoing for you’ ’ ’(P23).

Lack ofunderstanding .Abouta thirdof the non-disclosers (33.3%,9/ 27)referred to the possibility that other people wouldnot understand them :

‘Ijustfeel that people don’ t understand,and I thinkthat they’ ll know[there is something] abnormal aboutme if I tryto explainhow I feelabout my family’(P14).

Lack ofin terest orattention .Aboutone-quarter (25.9%,7/ 27)of the non-disclosing participants mentioned that they thoughtother people wouldnot listen or beinterested in their disclosure. Asone participant putit :

‘theymight not just be bothered or want to sitthere and listen to it’(P26).

Nop oint disclosing .Asimilar proportionof participants (25.9%,7/ 27)mentioned that they thoughtdisclosure wouldbe pointless. Forinstance : ‘Ijustdon’ t seethe point in tellingthem. I justdon’ t seethe point’ (P18).

Othercategoriesofanticipatedresponse .There were anumberof other categories of Shameand non-disclosure 9 anticipated response which were notedin less than20% of cases. These were lack of trust in others (14.8%,4/ 27),fear that the recipient will tell others (lack ofconŽdentiality), (7.4%,2/ 27)and fear of notbeing believed (18.5%,5/ 27). Themean numberof anticipated negative responses per participant was 2.8.These results suggestthat amajor factor in the participants’non-disclosure was aconcern about howothers mightevaluate them.

Shameandotherfactorsasso ciatedwith the self .Inaddition to ‘anticipated responses of others’there was aloose association of anumberof reasons whichseemed more self- related andintra-psychic (althoughin most cases these reasons coexisted with the more overtly social reasons for non-disclosure). These involved shame andother factors that seem to relate more toqualities associated with the self thanto howother people mightrespond. Much of the material includedin this section was elicited bythe question:‘Doyou think that if youtold somebodyyou would feel stupid,or silly,or ashamed of yourself?’

Shameandotherself-consciouse motions .74.1%(20/ 27)of the participants agreed that they wouldfeel shame if they told somebodyelse aboutan undisclosed emotion recorded in their diaries:

‘That’s theshamepart of it,to tellanyone. Um, becauseI meanI wasalways like somebody that they lookedat tosortanything out you know .Um, thesame now [inaudible] so. But you can’t makethem understandit’ (P38).

‘Um, Iwouldn’t, I wouldfeel I thinkmore shame than stupid. Um, Iwouldfeel that um, it’ s somethingthat I should,a problemthat I shouldkeep to myself,that I shouldn’t tellanybody about, um,’ cosit isshameful,it is a shamefulthing’ (P21).

‘Um. It’s not,uh. It’ s thisthing about men not talking about their emotions. Uh. It’ s, it’ s a,a girly thingto do. [Laughs slightly]. So I’ dfeela bitstupid. I’ dfeela bitashamed as well. I’ dfeela bit ashamedif Itolda man.I’ dfeeldeŽ nitely .Notso much if Itolda woman’(P4). Three participants whodid not agree that they felt shame useddescriptions that many shame theorists (e.g. Lewis, 1971;Retzinger, 1991)would regard as indirect expressions ofshame, e.g.‘ silly’or ‘uncomfortable’. Thefact that the majority of participants made remarks ofthis kindsuggests that shame was associated with non-disclosure of emotional experiences inthis sample. Guilt was seen as afactor in the non-disclosure ofan emotional incident by18.5% (5/27)of participants. As one participant putit : ‘IfI wasn’t feelingguilty because of thatI wouldhave explained and told someone’ (P33). Guilt was notspeciŽ cally probedin the interviews andtherefore may have been under-representedin the interviews.

Outo fcharacter .40.7%(11/ 27)of non-disclosingparticipants related non-disclosure to the fact that disclosure wouldsomehow contradict or underminea valued outward identity. 10 JamesMacdonald andIan Morley ‘I’vealways been really strong and in control. And to tr-,to, to then be reduced to relying on other peopleto help me, and, um. Other people usually tell me their problems, and to tellthem, it’ s too difŽcult for me . ..I’vekept it together up tonow ,but,I thinkit’ s justso difŽ cult, um, to putmyself ina positionwhere I’ mnotin control of thesituation’ (P8).

‘thefamily see me as a laugh,and a senseof humourand all that stuff, and, um, I don’t knowhow they’d seeme if Isaidanything’ (P2). Thecomments in this category suggestthat these participants are trappedin forms of relating to others which,while they may be validated byothers, are false to the participants’own emotional experiences.

Inability tojustify feelings andexperiences .37%(10/ 27)of participants referred to howthey felt unable to justify or account for their feelings orexperience. Forinstance:

‘womenof my age are, um, normally perfectly capable of having, um, a sexualrelationship with theirhusband. They are expected to have a sexualrelationship. There’ s noreason why they shouldn’t. For whatever reason I, Idon’t orcan’ t, um, and that is an extremely shameful thing. Um, Ishouldbe able to. Um, it’s, he’ s anextremely caring man, so there’ s noreason why I shouldn’t beable to’ (P21). Here the participants’experiences appear to beincommunicable because they contravene normsof behaviour andexperience.

Responsibility .33.3%(9/ 27)of non-disclosingparticipants mentioned that they considered themselves responsible for their difŽculties. Asone participant putit : ‘You’re onyour own. Y ouknowyour own problems. Y ou’ve gotto sortthem out yourself. It’ s easier saidthan done. It took a lotfor me to come here’ (P10).

Rejectionofone’sown feelings .18.5%(5/ 27)participants made explicit comments about howthey regarded their feelings as invalid. Forinstance:

’I’mnotsupposed to have these feelings, I’ msupposedto be self-disciplined,you know’ (P17).

Reluctance toe xperience unpleasantemotionsormemories .Over aŽfth (22.2%,6/ 27)ofthe non- disclosing participants made comments tothe effect that they didnot wish to disclose because it wouldhave intensiŽed or reinforced the pain of their feelings. Forinstance:

‘IfI don’ttellanybody ,peoplecan’ t remindme about it. And then eventually ,whenthe memory gets distant,you can blank it, forget it happened,you know, pretend that it didn’ t happen,or it didn’ t happento me’ (P16). Thiscategory ,similar to the notionof ‘emotional avoidance’in cognitive therapy,is the only one of the ‘factors associated with the self’that may bedistinct from the overall theme ofhowthe individual Žtsin with others.This is because reluctance to experience a memory oremotion could besimply because it is painful per se,andnot because of the social implications of experiencing it. Shameand non-disclosure 11 For‘ shameandother factors associated with the self’the mean numberof categories per participant was 2.2.

Isolation.Inaddition to the ‘anticipated responses ofothers’and ‘ shame andother factors associated with the self’, over aquarter (29.6%,8/ 27)of participants said that one ofthe reasons they didnot disclose was because there was simply noone available that they felt they could talk to.For example:

‘Therewas no onearound to tell. Apart from my husband, which I’ drathernot discuss it with’ (P24). Commentsof this kindseemed to underline the participants’sense ofisolation andlack of solidarity with other people.

Participants whoo nlydisclo sed .Thequalitative analysis reported above is based onmaterial from participants whofailed to disclose at least one emotion andwere therefore asked questionsabout why they hadnot disclosed. Seven participants (20.5%,7/ 34)only reported emotions that they subsequentlydisclosed andso were notasked these questions.This raises the questionof whether these disclosing participants’pre- occupations andbeliefs regardingdisclosure were entirely different to non-disclosing participants orwhether they were similar anddiffered chiey with regard to outcome (i.e. they disclosed in spite of their reservations); 57%(4/ 7)made comments which suggested that they disclosed in spite of experiencing similar pressures to withholdas participants whoremained silent. Thefollowing comment made bya male participant whose emotions were related to ashbacks of childhoodsexual abuse illustrates the kindof concerns these disclosing participants mentioned even thoughthey did conŽde their feelings in someone else:

‘There’ve beentimes when I’ veactuallywritten the  ashbacksdown and I’ vehiddenthe paper, so she couldn’t Žndit like,because I didn’t knowhow she would react to whatI’ dputdown’ (P6). Commentsof this kindtended to bemade bythe participants whoelaborated more in their interviews so it is possible that the remaining three participants experienced fears of this kindbut did not voice them inthe interview.

Discussion Following mention of the limitations ofthe sample, the main themes that emerged in the analysis will be discussed.Consideration will begiven to (1) the participants’apparent emotional isolation, (2) the nature of shame associated with the non-disclosure of emotional experiences, (3) the accuracy of negative interpersonal expectations about disclosing, and(4) the questionof whether,in spite of their tendency to keep emotional experiences private, participants actually wanted to disclose, if they were able to Žndthe rightcircumstances.

Limitations ofthe sample Thecurrent studyis the Žrst attempt to examine information onthe disclosure of speciŽc 12 JamesMacdonald andIan Morley emotional instances inany clinical sample. However, because the Žnal sample was only 27%ofthose invited to take part,and not all ofthose participants contributeddata to the main analysis (because they disclosed all the emotions they recorded in the diary), the degree to whichthese Žndingscan begeneralized to the broader populationof people referred for psychotherapyis severely limited. Thetentative conclusions discussed in the following sections shouldbe considered withthis inmind. It is hopedthat the Žndings can beveriŽed ina follow-up study.

Emotional isolation Perhaps the most strikingpreliminary Žndingwas the extent of non-disclosure in this sample:68%of emotions reported in this studywere notdisclosed, contrasting with a mere 4–10%of emotions incomparable studies withnon-clinical samples (Rime´ et al., 1991,1994). This tendency was underlinedby the theme of ‘habitual non-disclosure’in the interviews. Even those participants who only disclosed the emotions they recorded in the diary (N 7)more often thannot indicated that disclosing emotional experiences was problematic for them.

The nature ofshameassociatedwith non-disclosure Arich picture of shame andother cognitive-emotional factors related to emotional non- disclosure emerged inthe analysis. Mostof these themes could be loosely classiŽed into two broadcategories. Onthe one hand,shame was accompanied byanumberof factors which implied some kindof negative self-assessment. Inthis respect the ‘outof character’, ‘inability to justify one’s experience’, ‘ownresponsibility’ and ‘ rejection of ownfeelings’ categories, in addition to ‘shame’itself, all suggestedthat the individual judgedthemselves oraspects of themselves tobeunacceptable. Onthe other hand,many themes were related to participants’projections ofhowother people wouldrespond to disclosure inways whichwere unaccepting.Expectations that others wouldrespond by labelling, blaming or judging,that others wouldbe upsetor burdened,that they would disregard the individual byattempting to reassure them or give them advice (unhelpful positive responses), or that they wouldsimply lack interest or be unwillingto pay attention to them,added a strongsocial counterpart to the more obviously shame-related categories. Thisassociation of self-related andother-related appraisals suggeststhat the shame associated with non-disclosure is embeddedin broader interpersonal schemata relating to howa personexpects to be regarded andtreated byother people. Thisis congruentwith Lewis’(1971) notion of ‘superegoshame’ which includes imagery ofa punitive andjudging ‘ other’alongside imagery ofthe self as weak andinadequate. It also appears to reect the two facets of shame that Gilbert (1998)has termed ‘internalized’ and‘ external’shame. Thecurrent Žndingsresonate with the dramaturgical theory ofGoffman (Goffman, 1959,1963). In 1959, Goffman coined the term ‘destructive information’to refer to information which,if knownby others, would damage an individual’s attempts to present themselves in apositive light to others.He suggestedthat people are powerfully motivated to conceal suchinformation byadesire to avoid shame or embarrassment. The current analysis suggeststhat the burgeoningwork on impression management which Shameand non-disclosure 13 stems from Goffman’s theorizing andwhich tends to ignore shame (e.g. Leary,1995) could proŽtably rediscover the linkbetween the management of identity andthe management of shame. TheŽ ndingsalso echo Finkenauer et al.’s(1996)discovery that non-disclosure of emotional experiences was related to adesire to avoid shame, embarrassment andthe negative judgementsof others.The apparent pervasiveness of suchprocesses in this clinical sample is consistent with ahigherincidence of stigmatising life events andcircumstances (for example higherrates of CSAandhumiliating life events). It shouldbe notedthat, in addition to the broadly dramaturgical self andother- related categories, other factors may also have contributedto disclosure. Thecategory ‘reluctance to experience unpleasant emotions/memories’suggests that participants may have inhibited emotional experiences simply because they were painful andnot because of any identity implications of disclosure. Inaddition concern that disclosure wouldupset or burdenthe recipient andguilt also seem to standapart from dramaturgical concerns.

The accuracyof ne gative interpersonalexpectationsregardingdisclosure Theemphasis that participants placed onhow other people mightrespond negatively and unhelpfully to disclosure raises the questionof the degree to which suchexpectations are accurate. Ina review ofthe literature oninterpersonal aspects of disclosure, Kelly and McKillop (1996)suggest that, in general, people dohave alow tolerance for other people’s disclosures ofemotional distress. Thisimplies that the anticipation ofnegative andunhelpful responses byparticipants may bequite realistic. Interview material from the same sample relating to emotions which were disclosed (analysed in Macdonald,1999) showsthat very few participants reported disclosures which were met with anegative response,while over one-thirdreported disclosures whichmet with apositive response. Thisis consistent with the notionthat participants may have been reasonably judicious in assessing howpeople wouldrespond. Further exploration of this issue is important because it speaks to the degree to whichsuch beliefs shouldbe regarded as ‘dysfunctional’. Ina recent attempt to integrate cognitive therapy with ideas from community psychology,Hagan andDonnison (1999) have pointedout that many apparently dysfunctional beliefs, suchas ‘Iam worthless’may in fact be overdetermined bythe individual’s social world (which may for instance be sexist, classist, homophobicand racist).

The desire todisclose Finally it shouldbe notedthat, although the pattern ofnon-disclosure andemotional isolation seemed tobepervasive, many participants didexpress awillingness andeven a positive desire to talk abouttheir emotional experiences with others,given the right circumstances. Only50% (17/ 34)of the participants reported undisclosed emotions alone intheir diaries; 20.6%(7/ 34)of participants recorded only disclosed emotional instances andabout one-quarter (26.5%,9/ 34)of the participants reported instances of disclosure as well as instances of non-disclosure in their diaries. Analysis of the interview material relating to disclosure is presented in Macdonald (1999). 14 JamesMacdonald andIan Morley Acknowledgements Theauthors would like to thank Dr Stephen Gladwell for his help in arranging access to the study’ s participants.The authors are also grateful to Dr Elaine Duncan for her assistance in designing the diary protocol.

References Alexander,B., Brewin,C., V earnals,S., Wolff,G., & Leff,J. (1999).An investigation of shameand guilt in a depressedsample. British Journalo fMedicalPsyc hology , 72, 323–338. Andrews,B. (1995).Bodily shame as a mediatorbetween abusive experiences and depression. Journal of AbnormalPsycho logy , 104, 277–285. Andrews,B. (1997).Bodily shame in relationto abuse in childhood and bulimia :apreliminaryinvestigation. British Journalo fClinicalPsyc hology , 36, 41–49. Andrews,B. (1998).Methodological and deŽ nitional issues in shame research. In P .Gilbert& B. Andrew (Eds.), Shame:Interp ersonalbe haviour,psychopathology, andculture .Oxford:OxfordUniversity Press. Barrett,K. (1995).A functionalistapproach to shameand guilt. In J.P.Tangney& K.Fischer(Eds.), Self- consciouse motions: The psychology ofshame,guilt, em barrassment and pride (pp.25 –63).New York:Guilford. Brown,G., & Harris,T .(1978). The Socialorigin sof depression: astudy of psychiatric disorder in women . London: Tavistock. Drauker,C. (1992). Counsellingsurvivors ofchildhoodsexual abuse .London:Sage. Elliott,R., Fischer,C., & Rennie,D. (1999).Evolving guidelines for publication of qualitative research studiesin psychologyand related Ž elds. British Journalof ClinicalPsychology , 38, 215–229. Finkenauer,C., & Rime´,B.(1996).Emotionelle Geheimnisse :determinantenund konsequenzen(Emotional secrecy:determinantsand consequences). In A.Spitznagel(Hrsg.). Geheimnis—Geheimhaltung .Gottingen: Hogrefe.[English version available from Ž rstauthor]. Finkenauer,C., Rime ´,B.,&Lerot,S. (1996). Asocialm odel of secrecy .Paperpresented at the 11th General Meetingof the European Association of Experimental Social Psychology, July, 1996, Gmunden, Austria. Gilbert,P .(1998).What is shame?Some core issues and controversies. In P .Gilbert& B.Andrews(Eds.), Shame:Interpe rsonalb ehaviour,psychopathology, and culture (pp.3 –38).Oxford :OxfordUniversity Press. Gilbert,P .,Pehl,J., &Allan,S. (1994).The phenomenology of shameand guilt :anempirical investigation. British Journalo fMedical Psychology , 67, 23–36. Goffman,E. (1959). The presentationof selfin everyday life .London:Penguin. Goffman,E. (1963). Stigma:Notes onthe management ofspoiled identity .London:Penguin. Hagan,T .,&Donnison,J. (1999).Social power :someimplications for the theory and practice of cognitive behaviourtherapy. Journalof Community andAppliedSocialPsyc hology , 9, 119–136. Harder,D. (1995).Shame and guilt assessment, and relationships of shame- and guilt-proneness to psychopathology.In J. P.Tangney& K.Fischer(Eds.), Self-conscious emotions: The psychology of shame, guilt,e mbarrassment and pride (pp.368 –392).New York:GuilfordPress. Kelly,A., & McKillop,K. (1996).Consequences of revealing personal secrets. PsychologicalBulletin , 120, 450–465. Leary,M. R.(1995). Self-presentation: Impression management and interpersonalbe haviour .Madison,WI :Brown& Benchmark. Lewis,H. B. (1971). Shameand guiltin neurosis .New York:InternationalUniversities Press. Lindsay-Hartz,J. (1984).Contrasting experiences of shame and guilt. American BehavioralScientist , 27, 689–704. Macdonald,J. (1998).Disclosing shame. In P .Gilbert& B.Andrews(Eds.), Shame:Interpersonalb ehaviour, psychopathology, and culture (pp.141 –157).Oxford :OxfordUniversity Press. Macdonald,J. (1999). The Experienceofshameandthe emotionaliso lationof psychotherapypatients .Unpublished PhD thesis,University of Warwick. MentalHealth Foundation (1997). Knowingour minds: Asurvey ofhow people in emotionaldistress take care oftheir lives. London:MentalHealth Foundation. Miles,M., & Huberman,M. (1994). Qualitativedata analysis: Anexpanded sourcebook .London:Sage. Shameand non-disclosure 15 Mollon,P .(1984).Shame in relationto narcissistic disturbance. British Journalof Medical Psychology , 57, 207– 214. Oatley,K.,& Duncan,E. (1992).Incidents of emotionin daily life. In K. T.Strongman(Ed.), International reviewo fstudiesof emotion, vol.2 (pp.249 –293).Chichester :JohnWiley. Pennebaker,J. W.(1993).Overcoming inhibition :Rethinkingthe roles of personality, cognition, and social behavior.In H.C.Traue& J.W.Pennebaker(Eds.), Emotion, inhibition and health (pp.100 –115).T oronto: Hogrefeand Huber Publishers. Pidgeon,N., &Henwood,K. (1996).Grounded theory :practicalimplementation. In J. Richardson (Ed.), Handbookofqualitativeresearc hmethods for psychology and the socialscienc es (pp.86 –101). Leicester:BPS Books. Retzinger,S. (1991). Violent emotions: Shameandrage in maritalquarre ls .NewburyPark, CA :Sage. Rime´,B.,Mesquita, B., Philippot, P .,&Boca,S. (1991).Beyond the emotional event :Sixstudies on the socialsharing of emotion. Cognition andEmotion , 5, 435–465. Rime´,B.,Phlippot, P .,Finkenauer,C., Legast,S., Moorkens,P .,&Tornqvist,J. (1994). Mentalrum ination and socialsharing inemotion:Diaryinve stigations ofthe cognitive andsocialafte rmath ofemotionalevents .Unpublished paper. Sanftner,J. L.,Barlow,D. H.,Marschall,D. E.,& Tangney,J.P.(1995). Journalof Socialan dClinical Psychology, 14, 315–324. Sartre,J.-P .(1943/69). Beingand nothingness: Anessay onphenomenologicalon tology .London:Routledge. Stiles,W .B.(1995).Disclosure as a speechact :isitpsychotherapeuticto disclose?In J.W.Pennebaker(Ed.), Emotion, disclosure, and health (pp.71 –91).Washington, DC :APA. Tangney,J.,Burggraf,S., &Wagner,P .(1995).Shame-proneness, guilt-proneness, and psychological symptoms.In J. P.Tangney& K.Fischer(Eds.), Self-consciousem otions: The psychology ofshame,guilt, embarrassment and pride (pp.343 –367).New York:GuilfordPress. Tangney,J.,Miller,R., Flicker, L., &Barlow,D. (1996).Are shame, guilt, and embarrassment distinct emotions? Journalof Personality and SocialPsychology , 70,1256–1269. Tangney,J.P.,Wagner,P .,&Gramzow,R.(1992).Proneness to shame, proneness to guilt, and psychopathology. Journal of AbnormalPsycho logy , 101, 369–478. Wicker,F .,Payne,G., & Morgan,R. (1983).Participant descriptions of guilt and shame. Motivationand Emotion, 7, 25–39. Yin,R. (1989). Case study research: design and methods .NewburyPark :Sage.

Received 29November1999;re visedversionreceived16August2000

Appendix 1 Semi-structuredfollow-upinte rviewto thediary studyon thee xperienceand disclosure ofemotionalex periencesby psychotherapypatients (questionsif an emotionw asnotdisclose d) QUESTIONSFOR EACHEMOTION RECORDED ...2)IFEMOTIONNOT DISCLOSED:- (i) If youcan, can youexplain whyyou chose notto tell anyone?

(ii) Doyou think that if youtold somebodyyou would feel stupid,or silly,or ashamed of yourself?

(iii) What wouldhave to bedifferent for youto tell someone aboutthis?

(iv) Doyou think that not talking toanyone aboutthe way youfelt hadany effect on yourrelationships with other people? 16 JamesMacdonald andIan Morley Appendix 2 Example ofcoding

Excerptfrominterviewwith Participant4 INTERVIEWER:OK,well I’ll ask some more questionsjust aboutthis emotional experience, youwrote aboutin the diary.Um.The Žrst one is, if youcan, can youexplain whyyou chose notto tell anyone aboutyour feeling ofhatred? PARTICIPANT:There was, there was nobodyto tell. Theonly personI usually tell my problems to is my,uh,mother and,uh, Idon’t say everything to her.So, that was the reason really.Idon’t have, Idon’t have alot of friends, youknow .Idon’t have anybodyto talk to. Andthe friends Idohave, I,they wouldn’t be,uh. Well, people only want tohear so many things,and then they get fed up,so youdon’ t keep talking aboutthings like that. INTERVIEWER:OK.Doyouthink if youtold somebody,um,you would feel stupidor silly or ashamed ofyourself: PARTICIPANT:I’ dfeel stupid. INTERVIEWER:Stupid.Can you say abit more aboutwhat it wouldbe aboutthe situation that mightmake youfeel stupid? PARTICIPANT:Um. Its not,uh. It’ s this thingabout men nottalking abouttheir emotions. Uh.It’ s, it’ s a, agirly,girly thingto do.[Laughs slightly]. So,I’ dfeel abit stupid.I’ dfeel abit ashamed as well. I’dfeel abit ashamed if Itold aman.I’ dfeel, deŽnitely .Not somuchif Itold awoman. INTERVIEWER:And,what wouldhave to bedifferent for youto tell someone about this feeling? PARTICIPANT:W ell, they’d have to ask me. They’d have to ask me. INTERVIEWER:OK.Um,obviously ,you’ve told me aboutit, andthat was because I asked,obviously . PARTICIPANT:Y es. INTERVIEWER:Was there somethingelse that youhad in mind? PARTICIPANT:No, I couldn’t, I can’t really thinkof any ...I,Idon’t, I feel Iget drawninto conversations abouthow I feel, and,uh, its um,I get the feeling people ask for entertainment value. Not,to laughat, butjust ‘that’s interesting’you know . Butafter awhile these thingsare less interesting if youkeep goingon about them, they get boring.I tendto try notto. So Itry to avoid talking about[it most ofthe ?] time.

Coding for theabov eexcerpt

Habitual non-disclosure 4‘aftera whilethese things are less interesting if you keep going on about them, they get boring. Itendto try not to. So Itryto avoidtalking about [them most of the?] time.’ Shameand non-disclosure 17 Anticipatedresponsestodisclo sure

4(i)negative response (i)Believes other peoplewould get ‘ fedup’ . (ii)lack of interest/attention(ii) People ask about feelings for ‘ entertainment value’, butafter awhile‘ these thingsget less interestingif youkeep going on about them, theyget boring.’

Self-relatedreasons

4(i)shame, etc. (i)a) I’ d feelstupid’ , b)Psays hewouldfeel stupid andashamed because the feelings are ‘girly’. (ii)rejection of ownfeelings (ii) P relates rejectionof his emotional experiences to genderroles :‘It’s thisthing about men not talkingabout their emotions. Uh, It’ s agirly thingto do.’

Isolation

4Psays that hehasnobody to tell.‘ Therewas, there was nobodyto tell.’‘The only person I usuallytell my problems to ismy mother anduh, I don’t say everythingto her.So, that was thereason really.Idon’t have,I don’t havea lotof friends, you know. I don’thave anybodyto talkto.’ 18 JamesMacdonald andIan Morley Appendix 3 Datadisplay matrix for ‘habitualnon-disclosure’

Tableshowing datao nhabitual non-disclosure

P Interview material

2a)Although P hasrecently disclosed sexual abuse to acounsellorand a friendshe talks abouthow ‘ it’s alwaysbeen a darksecret’ . b)She talks about how shehas been ‘ likecovering up, so nobodyelse inthefamily clicks ontowhat’ s youknow,wrong’ . Shethinksafter counsellingshe’ ll be ableto tellpeople.

3 D

4‘after awhilethese thingsare less interestingif youkeep going on aboutthem, they get boring.I tendto try notto. So I try to avoidtalking about [it most ofthe ?] time.’

5a)P saidhe kepteverything to himselfbecause his abusers toldhim not to tellanyone andhe assumedthis was normal b)P saidthat ‘Inormallykeep, um, things to myself. ..’

6 D

7a)P says that nobodyelse needsto know:‘IfI toldanybody about it I wouldn’t forgive myselfreally .Because whatit’ s. IthinkI, Ijustwant to livewith it myown way ,and justlet me andthe wife and whoever else was involvedsort itouther own way. No, nobodyelse needsto knowabout it.’ b)P says that emotionalexperiences have no ‘bearingson anybodyelse’ : ‘Ichose,well thereason Ichoseis becauseI didn’t thinkit hadanything to dowithanyone else. It was somethingthat umI’ ddone.I was ashamedof what I’ d done,and that um [pause] andit was somethingthat Ihadto sort outmyself.And ithadno bearingson anybodyelse, and nobody needed to know.Of whatI’ ddoneand why I was feeling thisway .Sono,I didn’t Žndanyreason to tellanybody .It hadnothing to dowith anybodyelse. c)P indicateshis determination to keepexperiences associated withhis sexualabuse to himself:’It doesbecause I won’ttalkto, I willnot talk to anybody.Not evenmy wife’ .

8a)‘ Idon’t Žndit easy to discloseanyway.’ b)‘ I’maveryprivate person, and I Žndit really difŽ cult.’

10‘ Youknow, and I, IŽndit,I’ vegot friends,you know,but Idon’t discussanything like this withfriends, or .Andbasically I’ mjuston my ownreally, and I’ vegotto sort itformyself.’

11 D

12a) P says she’d Žnditvery difŽcult to talkto familyor friends:‘ImeanI, theyobviously knowthat I’vehad problems, but to breakit down and talk to them aboutemotions, but,I’ dreallyŽ ndithard.’(Though she says sheŽ ndsiteasier to talkto adoctoror a CPN). Shameand non-disclosure 19

b)P says ‘it’s notsomething that Iactuallyhave spoke about really’ .

13P talksabout how she can’ t ‘getit out’of her:‘[Inaudible]when I was sitting[inaudible] saidsomething to meas well. D‘‘Cos Igetangry .Igetangry ,butI can’t getit outof me.So Ijustsit there quietand take nonotice.Well, I pretendnot to takeany notice, but I doreally.’

14a) ‘ Ireallydon’ tfeelI wantto dragall the emotions up, um, with people, I justŽ ndit better leftunsaid.’ b)‘ Not, no,it justdidn’ t seem relevantto tellanybody .’ c)‘ Idon’t knowreally .Idon’t thinkI woulddiscuss it with anybody .Ireallydon’ t thinkI would.’ d)‘ wheneverI’ munhappyI don’t talkto other peopleabout it. I Žnditvery hard.’

15a) P keepsher intense feelings of guiltto herself:‘Idon’t talkabout it’ b)if anybodysees herwhen she’ s feelingthis way she says shehides it. c)P says ‘cos youcan’ t express itreally,so youjust keep it inside.By just keeping it inside.Um. ButI don’t Žnditeasy to share, that particularfeeling.’

16a) P isveryclear that shewouldn’ t talkto anyoneabout her CSA :‘NoI wouldn’t tell anybody’. b)‘ Icouldn’t see myselftelling anybody .’ c)‘ It’s notan option.’ d)‘ Idon’t wantanybody to know.That’ s whathappened to me.[Mm] Idon’twant anyoneto knowanything that hadhappened.’ e)‘ Yeah.[Y eah].The only way I canconceive of tellinganybody is if Iwas tellingit to them aboutsomebody else. [Mm] Butnot as me.’ f)‘ Ijust. ..Iwouldn’t betellinganyone.’

17 ? [says re ‘pointless’category that talkingto hiswifewould be ‘likebeating a deadhorse.’ ]

18a) ‘ I,Idon’t reallyshare ...outmy feelings and thoughts very much.’ b)‘ Isomehowkeep it allbottled up.’ c)‘ Idon’t reallydiscuss anything to, with other people.Not much.’ 19’ Ninetimes outof tenI don’t botherto say anything,because I justthink, you know ,he’s judgingme, on certainthings, by sayingthat Iam alousyparent to theextent that hesays I’vegotno controlover mychildren.’

20 D

21a) Said that shehad never discussed emotional situation with anyone apart from interviewand therapist. She said‘ Ifeelthat, uh,its likecarrying a loadthat youdon’ t wantto carry that,’ cos youcan’ t share itsometimes its unbearablyhard.’ b)‘ I’mnotopen, I’ mnotan open person.’ 20 JamesMacdonald andIan Morley

22a) Interviewer asks Pifshe is nervous about talking in the interview. She says :‘You knowto meits somethingprivate and I justdon’ t wantto share itwithanybody’ . b)‘ Idon’t know,Ijustthink its private.Why should I share it? Youknow,you know .I justbelieve leave it theway it is,you know.’ c)‘ as Ijustsaid before, its something,um, you just don’ ttalkabout. Y oujust try to keepit private.’

23a) P describesherself as a‘closedperson’ : ‘Um, Ithinkthat’ s justme generally.I’ m quite aclosedperson, and I don’tdiscusshow I feelabout anything, so itwouldbe naturalfor me notto tellanyone about anything that Ifeel[inaudible] . ..Itendto bottlethings up.’ b)‘ I’mjustso usedto nottalking about things, that Ijustdon’ tthinkI couldreally .’

24a) ‘ Idon’t thinkI’ d betelling anyone. No, no I don’t thinkI wouldhave done. Nothing canchange really . ..Ican’t imaginetalking to anyoneabout it.’ b)‘ Iwould.I wouldfeel silly and stupid. Y eahI would,very much.I’ maveryprivate typeof person.’

25

26a) ‘ theproblems I’ vegot at themoment, I’ drather keepthem to myselfthan . ..’ b)‘ Itry to hideall my feelings.’

27

28a) ‘ Um. IŽndit hard, um, to talkabout, sometimes Ithink,Oh its best that Ijust ignoreit, and I Žndthat wayI bottlethings up. And I try to blockit outmyself, but thetime I’m justgetting all worked up, but, sometimes Ithink,well its notworth tellinganybody .’ b)‘ Um, its justthe way I bottlethings up forso long,and I justkeep it to myself.Like anotherthing as well,being silly ,um,well perhaps to others itmightbe something trivial,you know, its nothingto worryabout. But to me,you know, it was.’ c)‘ Uh,its likeI say,[inaudible]I justtend to bottlethings up myselfand uh justtry andput itto theback of mymind . ..[Anddo youknow why you try to bottlethings up?]‘ That’s somethingI’ vealwaysdone . ..I’ve never done anything different.’ d)‘ ifthere’s anythingemotionally I don’treallysay anything,unless, you know, I have to ...Youknow I try to keepit, you know, a front,and just don’ t tell,you know, anybody.’ e)‘ I’vealways done this, and that’ s theway I do,I don’t realizethat I’mdoingit, but, sometimes Ichoosethat I’d, youknow, that Iwantto keepit to myself,I don’t want peopleto know.’ f)‘ Anythingto dowith myself I try,likeI say,Itry andbottle things up.’

29 D

30a) Before I turnedon the tape she said that writingdown and talking about personal thingswas somethingshe didn’ t likedoing —andthat was probablyanother reason fornot Ž llingin thediary . Shameand non-disclosure 21

b)‘ Yes, yes.I soundvery so-, I’mnotsolitary, I havegot friends. But I don’t goas far as theythink I dowith them, perhaps.’ c)(reluctance to disclosein interview) :‘Mm,because I grewup with the feeling of disgust,I meanintense disgust. And uh, that’ s somethingI try andput away from me,because it disgustsme aboutme too,although it hadnothing to dowith me. Um, butuh, I haveactually used the term Iam disgustedby something that happened.But again it’ s to dowith someone else’ s actions,which I foundnauseating, reallynauseating. Um, butoh, I don’t wantto talkabout that, ifyou don’ t mind.’

31 D

33a) P says sheŽ ndsit hardto talkabout CSA :‘IŽndit hard to talkabout it. [Y eah.] Because mymother didn’t believeme forso long.I Žndit reallyhard to liketalk to someoneabout it.’ b)P alsoseems to inhibitdisclosure of herexperiences more generally:‘Idon’t thinkI’ ll ever beable to tellpeople, you know. They’ d haveto belike, really ,reallyclose friendsbecause I don’t knowjust . ..Iexpectit’ s theway I’ vebeenbrought up, I’ ve neverbeen able to, if something’ s happenedI mean,you know, keep it to yourself. That’s howI feel.I’ d feeldreadful if I hadto tellanybody ,youknow ,anybodyand everybody.Imeanit’ s reallydifŽ cult to, like, to say umto work,I’ ve gotto goout. I meanthey, they don’ t knowwhat’ s happened,or what’s gonewrong.’ c)‘ I’mnotone for telling people how I’ mfeeling.I expectthem to readme [laughs slightly].’

35? [‘I wouldfeel it’ s notthe right thing to do’].

36a) ‘ Has ithappened? I reallyhaven’ t spokento manypeople. [Inaudible] professional people,even then I stillfeel, uncomfortable.’ b)‘ Idon’t know,yousee, I’vebeen very goodover theyears at buildingup brick walls.’ c)‘ It was awholegamut, gamut of things, and um, because uh, the guilt grew even more, andhas done progressively, um, it’ s somethingI ŽndIdon’t wantto share, to peoplewho, well I don’t thinkit’ s anyof theirbusiness [laughs slightly]. There we are. Idon’t reallythink it’ s um,if we’ re talkingabout colleagues or, or, or close friendsI don’t thinkit is,uh, I don’t thinkit’ s ofany value.’

37

38a) ‘ No,well I tendto coopit away . ..Itendnot to involvethem now,as muchas possible,you know, so. That’ s basicallythe trouble, I bottleit up.’ b)‘ Ijustkeep it to myself,you know.’ c)‘ Yeah,because I’ vealways bottled it up.I’ vealways sorted meownproblems out, you knowwhat I mean?’ d)‘ as Isay,I tendto bottleit upmore thananything, instead of tellinganyone.’

[D—indicates that the participant disclosed all emotions reported in the diary,andwas nottherefore interviewed aboutnon-disclosure ofemotional experiences]