FEATURE

How Children Tell: The Process of

Disclosure in Child Sexual Abuse

TEENA SORENSEN
BARBARA SNOW

III an analysis of ihe disclosure by a large number of

children from three to 17 years of (igv of huvin^ been
sexually abused, and who were ereniutilly eonfirmeil as
credible viciims. ihe large majority at first denied the
abuse. The authors describe disclosure as a process with
definable phases and characterisiits—seldom the sinf;le
entity thai typical investigations consider disclosure to he.
thereby setting the stage for ensuring injustices and harm
to the children.

Reported cases of child sexual abuse have reached epidemic proportions with
a 322% increase nationwide since 1980 (National Center on Child Abuse and
Neglect 1988]. The increase has raised both national consciousness and con-
troversy. A backlash movement has brought into question the credibility of
these reports, wiih complaints of false allegations and brainwashing |Crewd-
son 1988; Mechler 1988|.
Agencies nationwide have attempted to meet the demands foraeeountabiliiy
by developing methods and protocols that adhere to the rigors of the legal

Siircnsi'is. R.N.. M.A.. i.i ii Licensed I'.wihiiirrii Niirsi' .î/K'ciVï/i.vf in privtite

/viiciifi' iiiul II iloaoriil .•iluik'iu III ihe Univenity iifUkih, Sidl hike illy. UT. Bui hura
Siiinv. I).S.W., i.\ II IJccm-cd Cliniiiil Social Worker in prívale priiclire. Siill Lake
City. UT.

()O09-4(l2l/<)|/nilK101-l,-i $1 W (0 |g9| Child Welfare League of Amtrie;! 3

 4 CHILD WELFARE / Volume LXX. Number I / Januar>-Fcbniary 1991

truih-finding process, A critical part of the cicvelopment of such guidelines

should be an understanding of how and under what circumstance children
disclose sexual abuse. Pioneering clinicians have frequently asserted that
understanding the process of disclosure is an important element in assessment
of the problem |Dugan and MacF;irl;mi: I98K; Sgroi el al. 1982; Summit
19831,
Certain characteristics have been considered piirt of the disclosure phe-
nomenon. Denial has been identified as a frequent response when the child
is feeling too threatened, frightened, or insecure to acknowledge the abuse
ICourtois 1988; Dziech and Schudson 1989; Lister 1982; MacFarlane and
Krebs I986|. On the other end of the spectrum, one finds frequent references
lo recanting occurring for children who are pressured by family, offender, or
court proceedings |Faller 1988; Lloyd 1982; Sühd 1980|, Yet. objective re-
search findings to confirm Ihese clinical ohservalions are few.
Two types of disclosure are discussed in the literature |Sgroi 1982|; ac-
cidental disclosure—revealed by chance rather than u deliberate effort on the
victim's purt; and purposeful disclosure—when a child conseiously decides
to tell an outsider. Sgroi has suggested that accidental disclosures comprise
the majority of diselosures. Yet Saucier 1198^1, in a sample of 156 sexually
ahused ehildren, found a majority (.'Î.'Î'/Î) of the reporls were purposeful.
The breakdown of defensive barriers in the process ol disclosure has been
described as the "no-maybe-sometimes-yes" syndrome (MacFarlane and Krebs
1986|. Almost no data are available, however, in the literature confirming
this disclosure pattern or any other. Since the ahility to develop effective
guidelines and protocols musi rest to some degree on understanding this
y- phenomenon, it is critical that more descriptive data be made available.
/ This study examines the disclosure process in a sample of sexually abu.sed
children. It was hypothesized that these children would exhibit a disclosure
progression from denial to tentative to active disclosure. It was also hypoth-
esized that some of these children would recant, but most would later reaffirm.
The ;iutliors were also interested in pinp<>inting the type of disclosure as either
purposeful or accidental, and the motivation^ factors that prompted the
revelation.

Six-hundred-and-lhirty eases of alleged child sexual abuse in which the authors

had been involved as tberapists and/or evaluators from 1985 through mid-
Teena Sorensen I Barbara Snow 5

1989 were relrospectively analyzed. Eighty percent were seen at a large

mmprolii sexual abuse treatniont fycility, with the majority referred by police
and protective services. Twenty percent were seen subsequently in the authors'
private practice; ihis populaiitm was by private referral. The ehildren lived
in lour adjacent counties in eeniral and northern Utah.
Qualiiaiive analysis of clinical notes, conversaiions, audio- and videotapes,
and reports revealed common elements in Ihe disclosures of the 630 eases.
From these data, the proptised disclosure process was identified and variables
analyzed retrospectively in a subset of 116 confirmed cases of child sexual
abuse. The disclosure process was identified as containing four progressive
variables. Denial is defined as the child's initial slalenient to any individual
that he or she had not been sexually abused. Disdosurv is identified as
containing two phases, tentative and active; tentative disclosure refers to Ihe
child's partial, vague, or vacillating acknowledgment of sexually abusive
activily; active disclosure indicates a personal admission by the ehild of having
experienced a specific sexually abusive activity. Recant refers to the child's
retraction of a previous allegation of abuse ihat was formally made and
maintained over a period of time. Reaffirm is dctincd as the child's reassertion
of the validity of a previous statement of sexual abuse thai had been recanted.

Subjects
The majority of these 116 children were seen by ihe authors in a large
outpatient sexual abuse treatment clinic; the remainder were seen in a private
practice setting. They were referred by child protective service, law enforce-
ment, other mental health personnel and agencies, and private referral. All
116 cases were confirmed in one or more of the following ways:
1. A confession or legal plea in criminal or juvenile court by the
offender—80% of the cases;
1. A conviction in criminal court for one or more of the alleged
offenses—14% of the cases;
3. Medical evidence highly consistent with sexual abuse—6% of
the cases.
Of the sample of 116 children, one child was abused by two different offenders
in Iwo unrelated incidences, making ihe total number of disclosures 117.

Sample Dismbution
Figure I shows subjects ranged in age from three to 17 years. Twenty percent
fell in Ihe ihree to five year range; 4ü7c in the six to nine year range, 13%
in Ihe ten to 12 year range, and 27% in the 13 lo 17 year range. Sixly-lwo
 CHÍLD WELFARE / Volume LXX, Number I / January-February 1991

FIGURE 1
Age Kunge of Victims

6-9 10-12 13-17

Age (Years)

percent were female and 38% were male. Ninety-five percent were Caucasian,
with 5% all minorities combined, which is reflective of the geographical
area's minority population. Seven perceni were from upper and upper-middle
socioeconomic backgrounds, 73% from the middle, and 20% from the lower
soc i (»economic group.
The majority of the offenders were male (96%). Figure 2 sets forth the age
and sex of the perpetrators: 29 children were victims of adolescent mules;
four were victims of seemingly high-functioning adolescent females who
abused while caregiving; 77 were victirii.s of adult males; and one was abused
by an adult female in concert with her husband. Five children were molested
by men over 60; no children were abused by older women.
As to the relationships of offenders to victims. 58% of the victims were
abused by members of Iheir immediate family. The percentage rose to 74%
when members of the extended family were included. It is significant to note
that 979Í of the sample were abused by a family member or person in a
position of trust and supervision of the child. None of the children were
abused by strangers.
The sexual activities ranged from forms of penetration and fellatio to fon-
dling. For 10% of the children, the nexual activity was a single occurrence;
88% were engaged in multiple incidents ranging from several months to seven
years; two percent were unknown.
Teena Sorensen I Barbara Snow

FIGURE 2
Perpetrator Age and Sex

80
70
60 Male
Pemak
50
'S 40
30 29

20
10
1

Adolescent Adult Senior

Perpetrator Perpetrator Perpetrator

Results
Types of Disclosure
Table 1 examines accidental versus purposeful disclosure and reveals that
accidental (14%) was clearly the most common type.
Disclosure by age grouping revealed that school-age children showed no
real propensity to disclose either in a purposeful or accidental manner. Bui
comparing preschool age three to five and adolescent B to 17, using the
Fisher Exact Test, revealed a significant statistical correlation (p = .009).
Preschool children were more likely to disclose accidentally, and adolescents
were niorc likely to disclose purposefully.
The history of the child and family was examined to pinpoint as accurately
as possible ihe primary motivational factor thai caused the disclosure process
to begin. Table 2 lists all relevant motivational factors in the children who
disclosed accidenlally. Though the breakdown of dala by age shows some
interesting trends, ihe subcatcgory sizes are too small lo draw any statistical
correlations.
 CHILD WELFARE / Volume LXX. Number I / January-Febniary 1991

TABLE 1 Types of Disclosure

Axe
3-5 yrs 6-9 W.S 10-12 \r$ 13-17 yrs ,V
Accidental 21 37 12 16 86 74
(I7.2E) (37.4E) (1I.2E1 (2O.2E)
Punwseful 2 12 .1 12 29 25
( 5.8E] II2.6E) ( 3,HE) Í 6.8E1
No Disclosure 1 1 2 !
117 100

£ = etpecled.

TABLE 2 Impetus for Accidental Disclosure

Age
3-5 6-9 10-23 13-17 N %
Exposure to Petpetratur 3 6 6 9 24 28
Sexual Behavior 3 7 1 — II 14
Inappropriate S ta te me nl 9 7 — — 16 19
Bchuvíoml Sign 3 5 — — S 9
SharcU Ciin fide nee — 4 1 3 S 9
ID as a Victim 3 4 1 S 9
Confess i cm — 4 2 1 7 ft
Physical Sign — — — 2 2 2
Observai ion — — 1 1 2 2
Total 21 37 12 16 86 liJO

Exposure to perpetrator, meaning the child was known to have spent lime
with an alleged or convicted offender, was seen as the most frequent impetus
to disclosure in all age groups—28% of the population. Sexualized behavior
that was age-inappropriate and/or excessive was commonly seen in the pre-
school and young school years and was the impetus in K^i of the sample.
Inappropriate statement, such as "Suck on my pee pee. Mommy," again
was high in the younger age groups, which constituted 14% of the tota! sample.
A confidence was often shared by girls with a girlfriend who did not keep
the secret; this was seen with children school-age and above in 9% of the
sample.
The impetus to disclose purposefully, as seen in table 3, included educa-
tional awareness in 24% of the total sample and involved school programs
Teena Sorensen I Barbara Snoiv

TABI.K i Impetus for Purposeful Diüclosure

Ax.
5-5 6-9 ¡0-12 n-17 N %
EdUL'Htiiin Awijrcncss — 6 1 — 7 lA
Inlliiuiico !>!' ['tors — t — 2 3 10
Prosimily Ui Pcrpclratiir 1 — 1 1 3 10
Timu Discluhure 1 2 1 2 6 22
Angry — — — 7 7 24
Concern fur Other — 1 — — 1 3
UnkiKwn 2 — — 2 7
Total 2 t2 12 29 100

from touch prevention to divorce and social skills groups, with early school-
age children being the group most influenced. One child's disclosure was
prompted by a self-assertion discussion with her mother. Influence of peers.
10% of the sample, and predominately among teens, ranged from following
the example of another who disclosed, as well as encouraging and supporting
a victim to disclose through the reporting process. Proximity to perpetrator.
10%, included Ihc departure of the offender, which seemed to enable the
child to feel safe enough to disclose, or plans for an impending visit from
the otfendor, which increased the child's anxiety lo the disclosure point,
Timety disdosurc. IT^k. appeared to be situations where everything just seemed
to fall into place. For example, a boy's being bathed reminded him of the
sexual touching that had occurred earlier; his mother handled !hc discussion
calmly and supportively. and the child told htr aboul the sexual molestation.
An^er was a significant impetus for purposeful disclosure, exclusively in the
adolescent age group (24%).

The Disclosure Process

Figure 3 illustrates the progressive elements of the disclosure process. Almost
threc-founhs of all the children examined denied having been sexually abused.
Denial staternents were most commonly made (¡i) when children were initially
questioned by a concerned parent or adult authority figure; and (b) when
identified as potential victims and initially questioned in a formalized inves-
tigative process. Only 7% of the children who denied then moved directly to
active disclosure. Tentative disclosure thus became the common middle step
for the majority {lHWi of these children.
Table 4 lists the prominent features identifted in the tentative disclosure
phase, which presented particular concern for the authors because the children
 10 CHILD WELFARE / Volume LXX. Number 1 / January February 1991

FIGURE 3
Disclosure Process

Denial Tentative Active \ Recant Reaffirm

72% 78% 96% f 22% 93%

TABLH 4 Characteristics of Tentative Disclosures

Example
"\ forgot" Forgetting
"It happened to Joe"
Distancing
"It happened a long time agu"
"It only happened once" Minimizing
"He tried to louih me but I hit him and ran away" Lmpowemienl
"When he puti hi^ hands, ihere. I go tc) the pink forest"
"1 was only kidding"

frequently appeared confused, inaccurate, and uncertain, often vacillating

from acknowlcdgmcnl to denial. Figure 4 shows that approximately only one
out of every ten children (11%) sampled were able to provide a disclosure of
sexual abuse without denying or demonstratitig tentative features.
Active disclosure was eventually made by %% of the children, with two-
thirds of the subjects being currently abused. A strong majority (70%) gave
further inforniatioti over time about the sexually abusive activities. "Active"
disclosure meant that they were able to give a detailed, coherent, first-person
account of the abuse.
Teena Sorensen I Barbara Snow \1

FIGURE 4
Active Disclosure

10,0%
Unknown

79.0%' ''11.0%
Active Disclosure Active Disclosure
with Denial and without Denial or
Tentative Features Tentative Features

in approximately 22% of the eases, children recanted their allegations. In

several instances, they denied responsibility for their earlier disclosures, say-
ing that someone else (parents, therapist) had made them say those things.
Reasons lor recanting are shown in figure 5. Of those who recanted, 92%
realTirmcd their abuse allegations over time.
The time Iratne involved in the progression to active disclosure varied and
was unique to eaeh case. Some children moved from denial to tentative to
active in a single scs,sion; others took several months tu reach the active phase.
Discussion

Disclosure of child sexual abuse is best described by this research as a process,

not ati event. The common presumption that most abused ehildren are capable
of immediate active disclosure by providing a coherent, detailed account in
an initial investigative interview is not supported by these Undings, which
suggest that only 11% of the subjects were in active disclosure at the time of
the initial interview.
This raises concern because virtually all investigative protocols are designed
to respond to only those children in active disclosure. Issues of child sug-
gestibility, contatiiination, and false allegations have increasingly restricted
the amount of suppon that clinical, protective, and investigative .systems have
made available to assist ehildren through the disclosure process. A child's
 CHILD WELFARE / Volume LXX, Number i / January-February 1991

FKiURE 5
Reasons to Recant

1. Pressure from perpetrator

2. Pressure from family

3. Negative personal consequences

4. Videotaping

5. Retelling parents

6. Judicial proceedings

7. Investigatory police or CPS

initial denial, failure to provide immediate detail, or recantation may result

in Ihe dismissal of a valid complaint.
The use of terminology that acknowledges disclosure as a dynamic process
through which a child progressively moves, ralher than a single-outcome
eveni, could more accurately reflect ihe child's circumstance and improve
system response. For example, a child may describe the abuse but add an
illogical ending such as, "Then I punched him in the nose and ran away,"
or discouni a previous statemeni with, "I didn't mean what 1 said." The child
could be labeled as not being credible and the complaint dismissed as un-
substantiated or, alternatively, the child could be identified as in a tentative
disclosure phase that may warrant a continuing investigation.
Seventy-nine percent of ihe children in this study initiallydenied their abuse
or were tentative in disclosing it. Policies and procedures geared only to Ihose
children in active disclosure notonly fail lorecognize the needs of the niujority
but may actually place children at increased risk.
An illustration of this concern is the videotaping of child disclosures during
initial clinical or investigative interviews. Disclosure as a process suggests
that denial and tentative features may be prominent on early interview tapes,
which may then serve lo impeach a child who is later capable of providing
credible court testimony. An acquittal of the defendant may place the child
Teena Sarensen I Barbara Snow 13

again at risk but with fewer resources. If an informed decision directs video-
taping procedures, prosecutors and expen witnesses would seek to educate
judges and juries about the nature of the disclosure process.
Techniques that assess the credibility of a child's complaint by analyzing
components of the statement should consider whether the child was in active
disclo.sure at the lime of evaluation, and could be considered valid only if
the content of the child's statement accurately reflected what the child ex-
perienced.
Analysis of the types of disclosure, accidental versus purposeful, and the
various motivations for disclosure revealed that these elements appear to be
age- and development-related. The strong majority of children in this study
(74%;) disclosed accidentally. This most often occurred for all age groups
after concern was raised about the child's exposure to an identified perpetrator.
Perpetrators have been found to commit between two and five times as many
crimes as they were apprehended for |üroth et al. 19821. highlighting the
importance of identifying any child who has had significant exposure to a
recogni/.ed perpetrator and notifying parents of the potential risk. A child's
initial denial to parents or others should not be the sole basis of reassurance
that abuse did not occur.
Sexualized behavior and inappropriate statements constituted the next larg-
est categories under impetus for accidental disclosure. Together tbey account
for a full third of all such disclosures. These behaviors were noted only in
preschool and young school-age children, implying that younger children,
who have less cognitive awareness and impulse control, will reveal in an
unplanned overt manner. This finding stresses the need for increased aware-
ness by adults of the behaviors exhibited by younger children. Sexual acting-
out behaviors by children are seen as a direct indication that the child is or
has been a victim of sexual abuse |Sgroi et al. 1988|. The importance of not
discounting these behaviors prematurely as "child's play" but rather seeking
clarifying information is made evident by the data, especially with presch(K)l
children. They may not understand enough of what has (x;curred to make a
conscious decision to tell but will characteristically act out the trauma.
Purposeful disclosures took place in 25% of the cases, with adolescents
being significantly more likely than their younger counterparts to consciously
decide to tell, motivated mainly by angerat the perpetrator. Common thought
has held that the teenager who is restricted or punished may become angry
and falsely accuse a parent of abuse ¡Summit I9íí3|. but the data, in contrast,
suggest that the anger at parental restrictions serves to override inhibitions
and fears and becomes the fuel that drives the adolescent to disclose.
The influence of peers was also noted as a motivational factor prompting
disclosure by adolescents. Strategies that promote peer support for identifl-
14 CHILD WELFARE / Volume LXX. Number I / January-February 1991

cation and prevention of drug abuse might be modeled for similar efforts with
sexual abuse.
Among primary-school-age children whu purposefully told, tht: influence
of educational awareness programs was dramatic. The programs were not
exclusively on sexual abuse prevention but all identified inappropriate be-
haviors and stressed assertiveness and personal rights.
Children who disclosed sexual abuse in this study, whether purposefully
or accidentally, most often did so only after initially denying any such contact
(72%). This coincides with findings that most sexually abused childrï;n never
reveal abuse during their childhood |Finkelhor 1979, 1984; Russell 1983;
Silbert 1984). Denial by the children in this study was not expressed merely
as a passive refuting of involvement but ai times involved impassioned state-
ments where the child adamantly denied any contact. Indeed, the only person
for whom denial is as great as that of the perpcirator may be the victim.
When the abuse was revealed, it most often came tentatively (78%), cloaked
by the child's defenses. This phase is, therefore, a precarious time to try to
distinguish a tentative disclosure from a fulse allegation. Time may prove a
great decision-making ally, since the subjects in these valid cases progressed
almost unanimously (96%) to active disclosure. The time frame to reach the
disclosure varied greatly. Once children progressed (o thitt p<iint, however,
they were able to provide a consistent first-person account with unique detail
that substantiated the abuse.
Twenty-two percent of (he children recanted Iheir statements. Recanting
has become a recognized phenomenon in child sexual abuse cases, particularly
cases of incest, with research findings ranging from 12% to 33% |Cœwdson
1988; Faller 1988|. Of the children who recanted ibcir allegations, 93"^ later
reaffirmed the original conipluint of abuse, Tlicse findings may reinforce the
importance of viewing disclosure as a prciccss with phases that, like devel-
opmental tasks, can be resolved. Similarly, if Ihc phase is viewed as an
isolated event that ends or limits ihc extent of the child's support and contact
with the system, Ihen the child may become fixed at one phase and be unable
to resolve it or progress. Although not ¡ill victims experience all the phases
of disclo,surc, recognition of the continuum ciicouiuges mm effective case
management and intervention. •• 4

References

Courtois, C. Healing Ihc Intcsl Wnund. N y * Yiirk: W,W. Nonun, l9(iH.

Crewdsün, J By Silence Hclraycd. BuMim, MA: Lilttc, Brown, I9K8.
 Soremen I Bill hura Snow 15

Dugan. M,. and MatFarlane. K, Disclosure of Sexual Ahu^. In Response: Child Sexual Ahuso
the Clinical Interview, i-diled by K. Macf^arbtie and J. Feldmcih Neu Ycirk: Guilford, lyHK,
21-24
I)/iech. B.. and Schudson. C On Trial. Bcistcin, MA: Beacon Press. I')S9.
E-ailcr, K.C. Child Si:iiu:il Abuse. Nev, Y<.rk: Ciilutiihia University Press. lySB.
Finkelhor, D Sexually Vitiimi/ed Childri;n. New Yiirk: [-rec Press, IITI.
Hiiikclhor. D Child Sexual Ahuse: New Theory and Rc^',oarctl New York: Free Prcis, I9M4.
Groth, A. N,; Uingu. R. Li,; and McFadin. J, B, "Undetected Reciijivi^rn Among Rapisti and
Child Molesters." Crime and Delioquency 28 {1982): 450-458.
Hechler, D, The Batlle and the Backlash. lj;xingtc)n, MA: E-exin^ton Bixiks, 1988.
Lisler, E. "Forced Siltnco: Neglected Dimensions iif Trauma." Anieric;in iiiumal of Psychiatry
I.W, 7 (19K2): 872-t(7fi,
l.loyd, D. Tlie Currohiirinion of Sexual Viet i mi/.at ion ot Children. In Child Sexual Abuse and
tlic Law. edited by J, Hulkley. Washington, DC: American Bar Association, 1982,
Mac! arlane, K., and Krehs, S Techniques fcir Interviewing and Evidence Gathering. In Sexual
Ahuse of Young Children, edited by K, MacE'arlane and J. Waterman. Ne« York: Guilfcird,
198ti. 67-100,
National Center on Child Ahusc and Neglect. Study of National Incidence and Prevalence of
Child Ahuse and Neglect: I9K8. Washington, DC: U,S, Government Priming Office, 1988.
Russell, D. "The Incidence and Prevalence iii Entrafamilial and Hstra Tamiliai Sexual Abuse iif
Heñíale Children." ChilJ Ahuse and Neglect 7 (IW.l): Í3-U(S,
Sahii, D. Psycho log iLii I Assessment of Sexually Ahusing E-amilies und Treatment Implications.
In Sexual Abuse of Childreti: lmplication\ for Treatment, edited by W. Hcilden. Eingtcwood.
CO: American Humane Association, 1980, 71-86.
Sau7ier, M. "Disclo\un.- cif Child Sexual Ahuse." Psychiatric ClinicNof Nimh America 12 (June
IIN«»!: 455-469.
S|;rol. S.; Blick, 1. ; und Pciiter. F. A Conceptual Framework tor Child Sexual Ahuse. In
HandbiNik ot CImicul Intervention in Child Sexual Abuse, edited hy S. Sgrui. Lexingtun,
MA. Lexmgton Books. 1982, 9-37.
Sgnii, S.: Bunk, B.: und Wahrek. C. Children's Sexual Behavicirs and Their Relationship to
Sexual Ahuse. In Vulnerahle Populations, edited hv S. Sgrcii. lj.-xinjLton. MA: Lexinptcin
Books, IIBK, 1 24.
Silhert, N. Treatment ol E'rcistitute Victims of Sexual Assault. In Victims of Sexual AgjjrL'ssion:
Treatment of Children, Women and Men, edited hy 1. R. Stuarl am! J. (i. (¡reer. New York:
Van Noslrand Reiiihold. 1984, 251-269.
.Summit, R. The Child Sexual Abuse Accommodation Syndrome. Child Abuse and Neglect 7
I19K3): 177-19.'.
(Addre.s.t reque.fls fur a reprint to Teena Sorenseii, 1400 Foothill Drive. Suite #112,
Sidi Luke dry. UTH4I0S.)