Witness to Violence: The Child Interview
ROBERT S. PYNOOS, M D., M.P.H., AND SPENCER ETH, M.D.
In this paper, we present a widely applicable technique of interviewing the traumatized child who
has recently witnessed an extreme act of violence. This technique has been used with over :00
children in a variety of clinical settings including homicide, suicide, rape, aggravated assault,
accidental death, kidnapping, school and community violence. The easily learned, three stage
approach allows for proper exploration, support and closure within a 90-minute initial
interview. The format proceeds from a projective drawing and story telling, to discussion of
the actual traumatic situation and the perceptual impact, to issues centred on the aftermath
and its consequences for the child Our interview format is conceptualised as an acute
consultation service available to assist the child, the child's family, and the larger social
network in functioning more effectively following the child's psychic trauma.
 Journal of the American Academy of Child Psychiatry, 25, 3:306-319,1986.
Although there has been a growing awareness of the
importance of work with traumatized children such as
victims of physical or sexual abuse and kidnapping,
there is a larger population of children who have been
witness to violence, and suffer from the aftereffects of
that psychic trauma (Pynoos and Eth, 1985). For
example, the Sheriffs Homicide Division of Los
Angeles County estimates that dependent children
witness between 10 and 20% of the approximately
2,000 annual homicides in their jurisdiction.
It is difficult to imagine a more harro wing experience than for a chiid to witnes a parent 's murder,
suicide or tape. Anyone who has attempted to assist
children who have recently been so traumatized wilt
onderstand the difficulty in knowin how to roteed. The
chiid mat' exhibit mant' of t e c aractens ics o an acute
posttraumatic stress response. As a result, he ( * ;he ran
present ac minti nr muto Direct inqu ry ibout the traumatic
event mat' be unproductive, leaving the interviewer
feeling stymied and the chiid further entrenched in
de',achment. In addition, the chiid wilt frequently be in
a state of mourning for the lost patent, further
complicating the clinical interview.
In this papier, we describe an initial interview tech
Rece« rd .!0. 19a33: aeeepted April ~. 1983.
Dr. Pvnoos j Director. Program m Preuentwe lntervention i n
nique that has proven successful in heiping a psychiatrie consultant to engage
young children in conversation shortly after witnessing a traumatic or violent
event. It is intended for use witti children Erom 3 to 16 years of age. This
technique enables the interviewer to gain insight info the chiid's
onderstanding of the event and to characterize the behavioral and emotional
responaes in order to provide specific professional support to the mild reen
after the trauma. The interview technique has ondergors a series of revisions
as oor experience has groven, and particularly as we have teamel from the
children's oven comments about the interview. The interview bas proven to be
a gerent technique applicable for ure witti children who have witnessed
murder, suicide, rage, accidental death, aggravated assault, kidnapping, and
school or community violente. To data, we have employed it witti over 200
childen. ln addition, tor interview format has been readily taught to
other~iental health professionals who have themselves suceessfully used it in
a variety of clinical settings.
Tra4ma. Vu)lenee • and Bereauement ~n Childhc.w4, Dwiswn of Child P~~chuurv and . s.IL$tant Professor. Prvchuury and Bwo ehaviorai Sccences. L
'CL4- Netsmpsyehwtru Institute. 760 Westurood Pta.:a, Los Angeles, CA
90021, wherr reprinu inay be requested Dr. Eth is .4 sistant Professor.
Interview Format
Psvchiatry and Biobehauwral Sciences, C'CLk.Vesropsvcheatne fnstitute and
Ciinual .4ssutant Professor of Psychwtrv and the Behat'wral Sciences,
This specialized interview technique is designel for ure in rhe initial
C'nwersuy of Southern Caiifornfa.
meeting
witti a recently traumatized chiid. It is presented here as a coherent
The authws w sh to thank Drs. Theedon Shapiro, Robert Michels and E.
James Anthony for their :houghtfW ~-omments on oor werk, and Jar ue
three-stage proces: opening, trauma, and ciodure. The format
Berman for her elitemal and ;ecretarial support.
u beisra bv permittint the chiid fust to expres thé
~»u2 138/S612503 .4306 $02.0010 x;1986 by the American Academy of
Child Psychiatry.
im ct of th trau a
instaphor, bv ure of a Droiective free drawin, and story telling taak. This
opening enables the consultant to appreciate the child's preliminary meen of
coping and defensive maneuvers. Second, the interviewer shifts attention to the
actual traumatic episode. In order to Foster masterg of the traumatic anziety,
he , overcomes the efforts of the chiid to avoid and leng.
306
WITNESS TO V1OLENCE: CNILO INTERVIEW
and supports a thorough exploration of she child's
experience. Finally, she consultant can then assist she
child in erfdressing his or her current life concerns witti
an jncreased sense of security, competente and
rnasterv. As witti any clinical interview, she sugges order
may be modified somewhat as a function of she child's
particular responses, bot it is important to adhere to
she general format. Each major step in she interview
proces may be doplood through a series of drawings. In
out experience, she entire interview requires
approximately 90 minutes.
Prior to she interview, it is important to have obtained Erom she family, police or other sources some
description of she family circumstances, she violent ,
event, and she child's subsequens behavior or re sponses. The interviewer can then be alert to important
references or omissions in she child's account.
____________ First S tage: O pen in g
Establishing she Focus. After greeting she child in out
usual way, we share that we have had experience in
talking to other children who have "gone through what
you have gone through." Others can say that they are
interestel in onderstanding witti she child what it was
like to go through what he or she hes been through. By
making that statement, we establish a fixus tor she
interview, inform she child that he or she Is not clone in she
predicamenc~; and offer some ego support to she child
byour willingness to look logether at what hes occurred.
After these preliminary comments, we do not find it
necessarv or heietui to have other persons, e.g., family,
relatiees, or guardians present. We see each child clone
in a auiet room
Free Drcwing and Story T e l l i n g . Upon being seated,
she child is Biven pentil and paper, and askerf to "draw :
hatever you'd like bot something you can telt a story,
about." The child is reassured that she Quality of she
drawing is of no concern, and allowed to approach she
taak without disstadion or interference. By ~tepping
aaide she interviewer may entourage she child to
altend tune to she creative work. All she children have
spelled themselves to stils taak, although these may be an
initial period of hesitance. The youngest chil
dren, show onder 4 years of age, are engaged in play
along witti their scribbling and are askerf similarly "to
make up a story." Our emphasis is tor she children to
begin in whatever menner is moet acceptable to them.
This approach allows for she child's imagination to be
temporarily rollevel from reality and superego constraints (Waelder, 1933). The children we have seen
have energetically taken to this activity even hours after
witnessing a violent ovens. The drawings and stores
eert' considerably in their projective style and content,
from examples of nearly direct accounts, to richly
endowed works of fiction.
Children appear to be more comfortable witti thie
mannet of therapeutic engagement than she alternativo
style of direct inquiry. This opening is seen ultimately
to facilitate a later open discusion of she
traumatic occurrence.. The interviewer hopes by his
erpression of interest, level of enthusiasm, and occasional playfulness, to entourage the child to regain more
spontaneity. This step begins she proces of toontering
30?
she passive, lesschel stance of she traumatized witnes. The
interviewer entourages she child to elaborate further on
both she drawing and story. This can be done by general
questions, i.e., "What kappens next?„ or by inquiring about
some eedion or detail. Elaboration usually shows she
interviewer to gein an initial appreciation of she child's life
circumstances.
Treurneut Ref erewee. The key concept in this open
g ing stage is that she violent ovens remeins intrusive on
she chi!d's mind and wilt be represented somewhere in she
drawing or story. The interviewer's talk is to identity she
traumatic references.. These may be obvious or obscure,
bus are invariably present and recognizable.
White telling she story, she child is seen to struggle
witti unacceptable, painful or frightening feelings which
may disguise she traumatic reformco. The drawing and
story provide clues to she sources of she child's anxiety
and means of topmg.
We have found foor common psychologica) methode
in she ereadolescent groep to limit or regulate their and
witkin she first weeks after she evenj, enig in-f, an~tas
shows she child to mitigate painful reality by
imaginatively roversmg she violent oostome. The child
may provide a more acceptable onding. For exampie, one
5-year-uil whus'legei, a stunt man, was fatally shot in a
family feud,f ~uddenly introduced a safety net after a
clown in her story has been maliciously pushed off a high
wire. Another groep of children avoid renvinding
themselves of she event by eitint spontaneous thoutht jor
roetante, an 8year-old failed to mention she prominent
television set he pistod in she picture. When askerf, he
animatedly told of a program in which Bugs Bunny is shot
at bot safely outruns she attacker's bisets. These chiidren
display momentart' interruptions of their fantast'
elaboration in order to avoid associations to she trauma A
third groep cannot at first engage in fantast'. They remain
fixed to she trauma and only draw she aallel scone. They
will without request begin to glee an unemotional '
urnalistic bot incom toto account of she event.. For
instance, one 8-year-old introduced himseif by saying,
"I'm Tommy, my father trilled my mother," A fourth
groep remeins in _________________________ netent
state of anxious arousah-as if to propere for future
i
Y
308
R. S. PYNo08
Jn his story, one child emphasized his lack of
personal security afteer his father's murder. This •
year.old boy told in his story of a kidnapping Erom a
front yard that was once safe to play in. These children wilt
keep themaelven preoccupied witti thoughts of further
harm in lieu of discussing the real event.
S eco n d S tage: Traum a
?
Reisring the Experienee
Emotional Release. The transition in the interview
Erom the child's drawing and story telling to the ex•
putst discussion of the violent event is
a critical moment
for the clinician and child alike. We have found it timely
and practical to link some aspect of the drawing or story
directly to the trauma. For example, we might say: "1'll het
you winti that:" 1) "Your father could have been saved at
the end like the clown," or, ?) "Your babysitter could have
gorten away Erom the man who was about to slab her," or,
3) "By saying what happend over and over again, you
would gel used to it," or, 4) "Your father were stils here to
protect you." What often follows is a profound emotional
outcry Erom the child. vow the child needs to feel the
interviewer's willingness to be a supportive presente and to
protect the child Erom being overwhelmed by the intensity
or prolongation of the emotional release. The interviewer
must be prepared to share in the grief and horror and to
offer the child physical comfort.
Reconstrurtwr:
Catharsis does nor adequately describe the goal of
emotional release. Before the child proceeds to "relive the
experience," he first must attain a state in which he does
nor feel ton threatened by his emotional ren naes, a state
in which he hes the hope, at least, of being able to begin
to tops witti theet. íf this is adequately achieved, the child
will appear ready to provide a vetbal description of the
event. The interviewer can Lk:cn direct the child by
suggesting that "Now is a goud time to teil what happend
and what you saw." The child wilt rensafe the traumatic
milieu through various devices. He may first choose to
reenact in action or draw the violent stens, bat the interviewer must entourage the child to translate the actions
or pictoria! depiction info words. Props-dofs, puppets,
toys, wegpons, etc.-are made available. The child may
herome engrossed in the reenactment play, so that the
interviewer must be witling to participatefor instance,
acting as the assmient, victim, polsre or
rh
the tentral action the child witnessed when h asral
harm was inflicted: the push to the floot, the blow witti
the fist, the piunge witti the knife, the biast of a
AND 8. WFH
paramedic.
The child should then be supported in his focus on,
shotgun, the moment of forced *eiaal penetration. It
may require firm support by the interviewer for the
child to draw the particular moment of violente. Although a marked increase in aniiety may precede the
child's doing so, afterward the child appears
strengthened in his or her mastery of the trauma.
Occasionally, it is nor until thi8 step that the pain of
the reality is experienced. Again, the interviewer may
tactfully facilitate the emotional release by stating, for
instance, "I bet you wiek the gun had been pointed as you
drew it, away from your Eether, so it would have gnoe off
harmlessly."
Perceptual Experiene. We follow this description of the
action by addresaing the child's sensory experience of the
episode: the sight and sound of the gunfire, the screams or
sudden silene of the victim, the first sight of blond, the
splash of blond on the child's own tinthese the death egony
of the victim, and the sirens of the pokte arrival. This recall
cao be elicited by a comment such as "Boy, you must have
gorten blond on you." In several cases we have been
surprised to have the child add that he or she was wearing
the verg pants which had been stained witti blond during
the violent episode. In addition, whenever a child describes
an intense feeling state, we ank him about the concomitant
physical sensation. For example, after a child said "It feit
awful," we asked where he feit it. He replied, "My heart
hart, it was beating so Eest."
Throughout this account of the traumatic event, the
child's selection will be influenced by cognitive development and style, previous history of trauma, violente or
losse and the actual circumstances. The child is also
continually attempting to tops witti the accompanying
affects: helplessness, passivity, fear, rage, roofanion, vilt,
and wen sust ment.
The role of the interviewer is to function as a holding
environment in order to provide a safe and
ro tel setting so that the child can further work at
mastery lespits the rising anxiety level. The interviewer
does nor allow the child to digress Erom this allimportant taak. He may need to question the child to
ensure that the circumstances and aftermath are fully
reviewed. Following completion of the child's account, the
interviewer must be sensitive to how physically
exhausted and emotionally spent the child may be in
contrast to the asaal psychiatrie asanion. Relaxation time
and snacks should be offerel The child needs to feel that
he or she is beig adequately ratel for during this
emotionally challenging time.
Special Detailmg. The child may imbue a particular
detail witti special traumatic meaning (Freud, 1965).
These details are of psychodynamic importante and often
provide clues to the child's initial identification,
309
wITNESS 1'o VIOLENCE: CHILD INTERVIEW
J
, .
for exatnple, with she aggressor, she victim of, Wg may
add, she protector, including she pokte. One adolescent
girl became preoccupied because her mother had been
shot while wasring a drens she daughter had lept she
mother that morning. A 5-year•old dwelled'on whether
his deceased mother's lege were braken, in part because
he had worp leg casts as a toddler and wished to have her
Eiwed up in that way, too. Another boy painfully recalled
having been immediately made to weer she belt used by
his father to beat his mother to death in order to hide she
evidente. White recognizing she unconscious significante
of these details, she immediate goal of she interviewer is
to hel she child distinguish himself from either she victim
or she assailant.
Worst Moment The interviewer cao then proceed to
ank about she worst moment for she child. It may not be
what an adult would asname, nor even something as yes
mentioned. Even young children have sufficient
observing ego to nettact on she avant and then movingly
describe a uniquely painful moment. This may meiode a
memory from eerlier in she day or from she violent
occurrence or from afterward. One 8year-old broke down
in teers as he told of she moment when he found a razor
blade ander his suicidal mother's bleeding arm. He had
been sent out that day to boy she razor blades, he
thought, to make paper meeha objects. He cried in total
disillusionment. One 7•year-old girl painfully netstad that
her father had celled out her nickname as he died from a
gunshot wound. A 14-year-old girl described a moment of
intolerable anger when in passing her father at she pokte
station he leid, "I'm sorry," having jast shotgunned her
mother to death. A 5-year•old expresled his int
dssappo.ntme..Y vhMt on rhr..rvmas E...+n Claus had
not arrived bat a bad man, a killer, instead. This exchange
is a particularly empathie rooment for she child as he feels
especially understood and close to she interviewer.
Violente/Physk& Mutilation. The interviewer must
now be witling to guide she child to approach she impact
of she violente and physical mutilation. Children may be
haunted by an unforgettable visaal image and may
struggle to unburden themselves of she light. Certain
children may insist on drawing a picture of she mutilated
or wounded panent. Other children may be more
reluctant, bat witli proper support wilt draw she
horrifying and painful light.
We have been impressed witli she child's naad to
nastore an image of she panent as physically intact or
undamaged. In cases of perental death, she funeral offers
an opportunity to view she panent as once again
nastond. Inquiries about skin ceremony are especially
fruitfut. We wilt allo ank if she children have a pho
tograph of thee paraos and, if one ia available, are wilt
look at it witli tbe child during she interview. In addition
to eiding she grief proces, this step helpe she child to
invoke eerlier, kappier images of she panent to counteract
{.Ili7b
she more recent, gruesome light. Hoorever, she validation
of eiternal reality and physical death or injury nawis to be
confirmed and, especially, for she younger child to be
concretely represented in play or drawing. Only when
children are secure in she belief of their perent's pbysical
death have we neen children speek openly of their grief
(Furman,1974).
Coping witli she Ezperience
Issues of Hwnan Accouritability. These are acts of
human violente, not of naturel disasters. Struggles 'over
human accountability add considerably to she child's
difficulty in achieving mastery. As recognized witli adults,
,posttraumatic stress disorder ia made more lavare and
longar lassing when she stressor is of human design,
especially human-induced violente ( DMS-III; Frederick,
1980; West and Coburn, 1984). As one 12-year-old leid,
"I'm mad at she way she did go. Beaune that hart. I jast
wanted her to die naturally, not die because someone shot
her.„ The child must confront his awareness and conflicts
over who is responsible. He may wish at first to avoid she
issue by telling she avant an accident, bat this provides
only superficial relief.
The interviewer ezplores she issue of whom she child
holds accountable for she act, his own onderstanding Qf
she motive, and she trild's conjecture about ways it could
have been prevented. For example, we ank "Hoor coma
it happened?" And then, "What would make someone do
something like that?" 1f she assailant is a strenger, it may
be easy for she child to assige .~e diama. Huwever, f& iiiy
ViOiel ;e cao throw she chlld info an intense conflict of
loyalty and he may suppress certain thoughts as
unacceptable to other family membars. The child might
have already ezpressed his view
in she story telling. A •y
7
h e r
f a t he r
m o t h e r
f a t h e r
e v il
as
h e
f r o m
a t
b oy s
s c u f fl e d
f ur t h er
f a al t .
w h o
wi t li
p h y si c al
H o o r ev e r ,
i n
u n e x h e c te d ly
h e r
cl e ar ly
mi sl e d
h e r
f a t he r
n ei t h er
c o ul d
s h e
a c cu s e
d u ri n g
s h e
m e n t
s h e
t ou rl a
t h o u g ht s
m o t h e r
f o r
a s s i
ab o u t
w o ul d
b e
o f
sh e
s o
t w o
a n d
h er
s aw
h er
b r o t h er s ,
Ev er y o ne
s t or y ,
a p p ea r
b a t
T h e
h er
a sa al t .
S h e
r e s p o ns i bl e
e a r - ol d
a n d
t h i s
a r e
w ould
w h o
e l s e
c hi l d
i n t en s
n o t
m o th e r .
a d d
S o
m o t h er
s p o o t
w er e
h el d
a s s i g n ed
o n
t o
b l a me
t o
h e r
si le n t ly
p r o f e et
a bu s iv e ,
ma n d e m o n
Ma n n a
s h e
t ryi n g
s h e
t o
s he
a lc o h o l cc
t w o
y ou n g ,
t r ou bl e m a ki ng .
m e m o ry
h el d
h er
o f
ki m,
b r ot h e rs
d ea t h .
r e sp o n s ib il it y
o f
s h e
is
n o t
i n t erv ie w .
v ic t i m 's
p r ov o ca tiv e
a c ti o n s ,
as
a lw ay s
T h e
f o r
c h i ld
f i x e d
a nd
m a b e
n e e n
m a y
b e
m o s t
c o n t a nt e ,
in
w o n de r mg
t o
v ar y
d i st ur b e d
w hy
by
s h e
310
R. S. PYN009 AND S. E'TH
to have yelled at her eeltangel busband the following: eipression of these particular fantasies may provide
"Go abeel, shopt me. Show the kilo what a big man you enough emotional relief that certain children, who had
been teloctant to describe the mutilation, wilt now do so.
are."
One 8-year-old boy enacted a sequence in which he
Innen Plans o/ action. We have been particularly
imagined his Eether ia taken by ambulance to the
hospital, operatel on, the ballet removed, and the wound
impressed by she children's immediate efforts to te
verse their hel lessnees b formulatin a lan of action
that wound have remedied the situatien., Liften (1979) tien
referred to these cognitive reeppraisals after catastrophic loss of life as "inner plans of action" and
suggests they are repeatedly asel to contentwith the
Jdeath imprint." As we have observed, the inner plans of
action may soek to alter the precipitating events, to
undo the violent act, to naverso the lethal conse,quences,
or to gein safe retaliation. Their content and time frame
are developmentally influenced (Eth and Py noes, 1984).
Because of their limited cognitive skilis, the prel
school children do net appear to imagine alternate
actions they might have taken on their own t.o erevent or
alter the episode.,It is they, therefore, who faal the most
heiplas. They may choose to flee or stat', look or turn
away, be attentive, or try to sleep, bul all of these are the
choices of a passiva witnes, net a participant. The
ereschool chiid may fantasize about potsilo help having
provided needed third party intervention, and, in his play,
look to the interviewer to fi11 this role.
In contrast, school age children do net act as mare
witnessen. They can participate, if only in fantast'. They
can imagine having celled the police, lockin vh doorn,
grabbing a wegpon away from
assailant,
mother's lege ware braken also had the opportunity to
ank a doctor for help loting his story play. Witti the
interviewer's aid, he then fixed up the braken logs of
aid to the victim. Not ismgly, these innereven
plans
capturing the
of action are n
always confined to assmient,
the day ofrein
the
y o ering
ava nt. rFor e xam pie, witti toe intervicwcr
p!aying the
role of the assmient, ene 10-year-old boy acted out how
he imagined surprising the manlener, kicking the gun oi".
of his hand, and lossing it to his unarmed Eether.
Adolescente can imagine alternativo actions over a
mach longen eerled of time. They do net merely fan
tasize participation, bot can implicate their own action or
inaction in a more realistic fashion. One 14-yearold
bitterly regretted her Peilure to ampel her father's gun
when she had had the chance 2 weeks before. A 1?year-old boy, in trying to stop the tape of bis
mother, was overpowered by the assailant in a hand to-
hand knife fight, and, afterward, continued to imagine
ways he could kill the man if he ever faced him again.
Especially important are ways the chiid imagines the
panent could have been "fixed up" or aided, _ the injuries
ware sustained. For example, the young boy who was
concernel witti the thought that his
his fictional injured race can driver. Furtherrnore,
stitched closed. Witti encouragement, he then readily draw
the view of the bloeding chest wound that continuel to
intrude on his mind.
We carefiiily explore all these cognitive reappraisals
(Folkman and Lazarus, 1984) for they are the best
i,ndication of the wars the chiid is troubled by feelings of
self-blame for net doing more. Enactment of these "inner
plans of action„ can offset lingering feelings of personal
responsibility.
Punishment or Retaliation. This discusion of blame can
raise the question of punishment or retaliation. It may
prove diff colt for the chiid because it can reveal
unbearable Peelings of Built in neme children and
frightening fantasies or dreams of reven ge in ethers. In
part, these Peelings serve to counter the true helplessness
at the moment of the violent act. We wilt allow the
children to Biva Eidl expresion to these , Peelings
_______________ remindias theet of the realistic limitations
to w t tinut couId have doge at the time. , The children
effen look relieved by permission to imagine the tortures,
multilations, or execution they have reserveti for the
assmient, and readily draw a picture of "What you'd like to
nee happen to him." Afterward we will raspond, "I nee it
Pools goud to imagine Batting back át the had man who
stabbed your father," adding "1 meen, to be able to do
something to him now, when you really couldn't have
stoppel him at the time."
Counterretaliation. Themea of nevengo may be associated witti fears of counterretaliation by the eengilent.
The chiid may be afteil of the assailant's return and
confused over what hes actually happened to the suspect.
If the assmient is already arrested, the chiid may be
fearful over the fotore release of the suspect. We are
concernel about how rarely the chiid is raassured about
these matters by the police or the local district attorney.
Child's Impulae Control If the chiid attributen the
assailant's action to enger, bate, rejection, om craziness,
etc., it is pertinent to ank, for example, 'What do you do
when you Bet angry?„ and to ex lore the challenge to the
child's own impulse control. Viewing an open display of
violence may net only causa the chiid to ipso trust in adult
restraint, bot he or she may acutely foet his or her own
capability, especially in
Í
WITNESS To VIOLENCE: CHILD INTERVIEW
Light of coneciouz revenge fantasie, and be concernel
about the lack of proper eiternal support.
Previous Trauma. After this discusion, it ie comInon
for a child spontaneoualy to menden pestlno • etstic
experiences. We have leernel from children of further
inlaates of child abuse, violent family deelha, unreported
suicidal behavior, physical injuries, or accidenta.
Traumata Dreams. At this point, we inquire about
recent dreams that may be remembered. Often a child
sepotla anxiety dreams directly teleled to the traumatic
event. The child may be fearful that the dream
representa a portent of the fotore, nol only of being
victim of a violent aasaalt, bot sometimes betoming an
evengrog killer, toe.,
Foton Orientation. It is now appropriate to aak the
child about his or her concerns for the fotore, specifically as they relate to the potential langers in interpersonal relationships. The child may have immediately
after the trauma crystallized a _v_ivid and re-, ~tricted
view of his or her own fotore. One child described how
when he grows up he intends to live in an unaccessible
fortress surrounded by many gaard logs. Mant' children
slatel that they never latend to many or have children
for fear of a similar violent outcome. Even school age
children sometimes described changes in career plans
for when they got eides. For example, one 7-year-old,
withm days of her father's killing, sepottel she suddenly
decidel on a new life ambition, to become a stand-up
comic who dressed in tags and made ethers laugh.
Similarly, the child feels burdened witti an awareness
of his or her unfortunate legact'. Children will complain
about the novelty and stigma of being the heit of a patent
who lied by murder or suicide. For lastante, one 11-yearold girl bitterly lamented her face as a laagtiter of a
'man who barnel himself to death.'
Current Stresses. When sufficient mastery of the
traumatic anxiety hes been achieved, the child can more
actively address the life stresses engendered as a
consequente of the treurneut avant. He or she may
spontaneously and pointedly inquire about placement or
schoolmg, or be easily encouraged to do so.
We survey a number of the temmen, easily overlooked
issues that may add to the child's distraas. These rotlade
tentacts witti the pokte and legel system, changes in living
situatien or schoeiing, awareness of media coverage, and
concerns about social, sti sta. We offer to help redres any
oversights in the child's care. For example, in gelag to stat'
witti her grandparents after her mother's laatdar, a 14year-old had to abruptly change schools and lost the
companionship of her established tinla of friends. We ware
311
able to asrange to have her old friends visit her at her new home.
Exploration of these poettraumatic consequences enables the child
to consider the impact of the event on hie tomaat life
circumatances.
Third Stage: Closure
Recapitulation. The sensitive pretera of terminalmg the
interview is now begon. The first step is to elicit the child's
cooperation in re ' and summarizing the wasion. We
attempt tó make t e child's responses weet more acceptable
by emphasizing how onderstandable, realistic, and
universal they are. By doing so, we also hope to have the
child feel leas alen and alienated, and more ready to
receive further support from ethers. The interviewer
returns to the initial drawing and story.. The link to the
trauma may be more clearly indicated perhaps by pointing
out a similarity to the child's later reenactment or
recounting.
Resustic Fears. It is important to repeat that it is all ri t
to have feit helpless or afteil at the time and then rad or
angry. We also make reference to what ether children
have teil us , after being in similar circumstances. The
threat to the patent is so overwhelming at the time of the
violente that many drildren do nol entertain a realistic
appraisal of their own personal jeopardy. Afterward, they
may ignore, leave unacknowledged, or suppress any fear
they might have experienced for their own safety. In one
case, we could point out how far away from the shooting
the child piscel himself in his drawing when in fact he
was so close as to have easily been shot himself. This
intervennen alleviated rather than aggravated the child's
anziety, perhaps by unburdening the child of the need to
suppress his fears.
Ezpectable Course. We share witti the children the
expecteble tourre, for theet as they pass through the
tourre of their traumatic netflens. For ezample, we might
say, "There grill be timer at school when you think about
year mom, and faal rad." Or, 'You may faal frightened to
ree a knife at home.' Or, "You may jump at the sound of a
lood noire," or "You may have some bad dreams bot they'll
happen leas and leas witti time." We suggest they share
these restbons witti trustel adults to gein further
assistante at these ditficult moments.
Child's Courage. The child's beleaguered self-asteem
nawis support. We may be able to acknowledge the child's
bravery. For instance, in one case a 5-year-old stamparel
out a window and down a fire escape to reek help for her
woonled mother. Qne convenient method is to retlect on
the child's performance dunne , the interview, net only
telling him what a goed job he lid bot, more important,
complimenting him on his courage to engage in rotti
difficult talk. We will ac-
312
a. S. PYN008 AND 9. frH
tualy say, "You are verg brave." Children invariably
grew for me and my eister and cut info the ehape of a
tien and a hippe. Fm going to etska aura to keep
swell witti prils upon hearing these word.
watering theet.
Child's Critique. The child is then askerf to describe wbat
bas been helpfui or disturbing about the inter• view. The
Consultant (C) : This ia the kind of Story you weidl
children are usually quite candid. They wilt describe what have likel to have happen instead of what lid.
issue had been "toughest" to talk about, what had been
Lisa: Yes (she begin to cry and is comforted by the
unhelpful, and what made theet faal tetter. In fact, they interviewer). She had jast been offerel a job a couple of
have been our best teachers. Before we understood days hefere the accident happend. She never had a
the rota of suppressed fear, ons child turnerf to ons of chance to take it. We could then have afforded to move.
us and eiclaimed, Boy, was it goud to say how afraid I Now she's moved to heaven and we're going to have to
was."
move somewhere eire (she continues to cry and accepts
Lesas-takin . As we end the interview, we glve expresion
physical consolation).
of our respect for the child and the privilege of haaing
C: i know it's going to maks you rad to talk about soms
shared the interview experience witti hun. We then of this.
emphasize our availability to be celled on in the future.
Lisa:1 don't care.
We generaily give the child, tinwever young, our
C: Maybe this is a goud time to teil what happend (she
professional card witti our telephone naether. It is then begins to describe the actual murder scans and was
important to issue open the opportunity for contact, as encouraged to draw a picture of it (Fig. 2)).
effen the trauma wilt be reactivated-on its anniversarv
Lisa: This is my mother handled up on the coach.
date, for instance. At such timer mant' chiln dren have She's jast waking up. The man is right here, talking in
sought us out lespita the brevity of this initial front of her face. At first I was jast taking a nap and the
contact. In these cases where the child hes been referred
baby was jast in tlie bedroom. When I came out and save
for further treatment we have observed that this hun witti the gun I was so scared, t jast stond there
consultation witti proper closure hes facilitated the looking. He was flicking the gun in her face. He then
child's adaptation to the treatrnent situatien.
shot her. He kapt shooting her until he got to the door,
and then he ren out somewhere.
Case Illustration
I waited until I haard our screen door shut, `causa I
Lisa, who is 11 years old, was interviewed 5 days
didn't
want to gat shot. Then I ren up to my mother, who
after her mother had been fatally shot by the rnother's
was
rolling
off the coach and she feil on her aide. I rolled
estranged boyfriend, .Jim. On the day of the murder Lisa
her
over
and
was talking to her. I was crying and I was
and her mother veere at home babysitting for a neighbor's
mal
because
I didn't know what to do. I grabbel the
infant, white Lisa's younger sister was at school. When
phone
to
call
the police bat I think they'd already been
interviewed, Lisa and her sister veere temporarily residing
celled
by
the
neighbors.
Before the police came I was on
witti a cousin and her young children.
the
floer
witti
her.
Her
eyes veere down, her evelids
The consultant began the interview by inviting Lisa to
veere like half-closed, and her hair was kind of messed
draw a picture and maks up a story stout it (Fig. 1).
L..: We weke up in the morning and we veere up, sticking out. B1ood was everywhere. I was trying to
packing. Then we gat info our car and the moving open her eyes. I Was trying to wake her up-you know, by
truck. We're losding everything onto the truck and we're shaking her face. I thought maybe if I lid something
moving so that if Jim came back he wouldn't find us, so wrong she weidl do something. 1 trial to listen to her
he can't sheet my mother. That's my mother who is haart and to take her pairs. 1 couldn't find anything. I'd
moving the plant. My sister is watering the gras and reen that on a soap opera 1 watehed witti my mother
When somebody got shot. Also, you know how when
I'rn heiping to take the planfa off the porch. And
you can't breathe how you pres their stomach or
everyone is feeling happy.
something, I Was trying to do that so she could breathe,
My mother lover plants. She had house planfa and all
kinds of planfa. I don't really know the name of theet bat bat she wasn't breathing right, and when she look a
she had locs of planfa around the house. She look goed breath thee blond startel to coma onto the goor. I kegt
care of theet. She didn't really like us heiping witti talking to her bot she wouldn't talk. 1 thought if only she
the plants because we might do something wrong and could talk then everything could be all right. I jast look
they might die. She wanled to take care of theet her hand in mine, and kegt shaking it white telling her to
bang on. I was crying and 1 askerf God to save my
herself.
Mommy and I'd be a goed girl, bot he look her (Fig. 3).
Right afterward the plants veere all given to
ARerward i went and teckel all the doors.
neighbors ezcept two. These veere ones Mother
C: You tried your best to help her if you could. I'm
especially
313
wrrNESS 1'0 VIOLEHC& CHILD INTERVIEW
1
3
J
5
r
r
I
3/
r
r
FIGS. 1-5.
Patient's drawinp of traumatic event.
sure you couldn't stop the bleeding. You worry you
could have dove more.
Lisa: There was ton mach blond
C: Did you get blond on you, ton?
Lisa: It was on these pants. These veere the pants I was
weermg. gut they've been washel 1 guess I had some
right here `causa I was listening to her haart. Her shirt was
Eidl of blond, and when she trial to breathe she made a
sound like this (she inhaler in a gasping menner) and
she lid it for a long time. Sometimer I remamber how
the gun sounded, how lood it was.
C: It's really something to go through.
Lisa:1 am not really mad at the fact that she's deed
because you know everybody bas to go, bot 1 am mal at
the way she lid go. Because that hort. I jast wantel her
to die naturally, not to die of a shooting or stabbing or
strangling or something like that.
It juat doesn't faal right. Oh, how she lied. Like that,
horrible. She should have lied naturel, and maybe I
could have been groven because that way I wouldn't have
lost a mother truly. I was scared when the police came
because first they knocked on the door real lood But I save
the Eire truck and it had a eiren on. So I knew it was the
polka. They didn't beige in as I thought they might
have. when the polka came I wantel to stay with my
mother bot they moved me to the next room for awhile.
314
R. S PYHo08 AND $. ITH
u$ to do things their way. It'i kind of hard for ua becauae we jast loet a
Tbat's bad that hei reet been found yet. I vees
thinking tbat my mother would be looking for kim and
teil me, $o then I could telt the police and they'd Eind
kim (she draws her mother in heaven (Fig. 4)).
She is in heaven witti God. He kas long halt. Jesus
had long halt. She's looking down on us. When she
neen us cry she vries toe. Shell maks sure we've all lelt so if
.lira vemen back he'll have a big surprise because nobody
will be there.
C: Do you get scared?
Lisa: i got scared this morning. I didn't want to go
outside because Jim could be somewhere outside. I askerf
my uncie to gel kis gun and scans away the man i save
across the streef. Maybe we could catch Jim, gel kim to
drop kis gun and bring kim to the police station.
C: You valled it are accident, net murder.
Lisa: ,lust a name I picked for it. But a man really dia it.
Marden sounds like when somebody chokes or strangles
somebody. I guess I'd say he killed her because that
doesn't meen so much physical contact. Mavbe if I
haarei weke up she wouldn't be dead. He probably got
scared when 1 save kim witti the gun.
C: And he dia something terrible to your mother.
What maken a person do something like that?
Lisa: I guess hete and enger. He wasrei out friend and
she tolel kim she diarei want to nee kim. I guess he got
angry at that.
C: What would you like to nee happen to kim?
Lisa: I was dreaming that all my cousins and out
relatiees veere dressed, you known, how they put you up
against the Wall and blindfold you and sheet you. i had
the name knife he used to slab my mother and the
name gun that he shot her witti. Then i went up to kim
and raid, 'Do you remember this, now you care feel it.' And
i stabbed kim rght where he stabbed my mother. Then 1
nala, "1 guess you remember this, toe,' and then 1 shot
kim. Then 1 meeerf, and evenbody starterf 3hooting kim
(she comments on how goed a pic.ure that is to draw Fig.
5I ). It feit goud like I was getting back at kim. The hands
are hekma kis back. He's lied up and the scarf is
around kis neck because we look it off so he could nee
what kappens.
C: Do you know what will happen if he's arrested?
Lisa: Heil go to jail. But I want kim to stat' in jail for
the rest of kis life, since he look a life from somebody he
ought to gel kis life down to nothing.
C: There must be timen when you gel angry at
somebody.
Lisa: When 1 gel angry 1 don't like to show my
Peelings. I jast keep them in. I held my breath and jast
don't think about it. But since my mother's death I can't
really control it all because the way people treat us it
seems like it's all out faalt she was dead The way
they're telling us to change so quickly. They want
mother. He must have had a hot temper. When t gel maa at my eister
now 1 gel afraid i might hit or hart her. I don't like that.
C: Do you ever think about her relationship witti men
and what it will be like for you when you gel eiast?
Lisa: 1 nala I'd never gel maaierf and have kils, cauee if
me and kim light, something might happen to me where I
have to die. Then I will leave my klas jast like my mother
had to issue. And I don't want to do that to my kids.1 had
that thought the naroe day she died. I was thinking about
when I have groven kils she'll never be able to nee her
grandkids in person.
(Toward the end of the interview Lisa began to address
neme of her current concerns.) We might have to go to are
orphanage. I don't want to. Nobody is there like relatiees
or anything. One nessen I don't want to go to a losten
home is because they might separate me and my sister.
It's bad enough we lost out mother.
C: Remember you tolel me how veeli your mother looked
after her plants. I think you want somebody to look after
you witti that much care. You must feel that God is being
very unkind in letting this kind of thing happen in your
life.
Lisa: Yeah. That sure isn't fair because I understand He
wanled her to issue us in this way bul why dia it have to
happen? Or, the least He could have dove was walt until a
verlam age when we could have been on out oven.
C: How kas it been to talk witti me?
Lisa: It made me feel hart and stuff, bul it also made me
feel goed.
Overview
We have described a complete interview fermst designerf
to assist children who have witnessed a violent act. The
heiplens, passies experience of watching the moment or
moments of a violent act and lts injurious consequences
constitutes are immediate psychic trauma for the child
(Pynooe and Eth,1985). The eiswing is painful, frightening,
and distressing. Our werk bas been witti the moet
traumatic cases: a perent's suicide, murder, or raps (Eth
and Pynooe, 1983; Pynoos and Eth,1985; Pynoos et al.,
1981). We think the "hammen effect of examining
catastrophic violence kas brought to light more visibly the
processen needed to werk witti children in mant' ether
clinical settings.
Although the technique we have described mat' appear
rigidly structureel, in fact, the sessions generally fellow
the child's leed and are rarely experienced as
,
315
childhood was written by Levy (1938). "Release therapy"
was specirically devised as a paychotherapeutic technique
to resolve symptoma arising trom a "definitely koewo
traumatic episode." The child's traumatic anxieties,
especially these related to aggressive behaviors inhibited
by fear veere discharged through directed play. Levy's
technique hes net received further exploration, in part
WITNE99 To VIOLENCE: CHILO INTERv1EW
`rbitrary or stifling. It would neem that there is an
underlying logic to the interview procesding in this way.
We have been impressed that neme child psychietrists
intuitively eenduet their initial interviews witti
traumatized children in exactly this way without consciously recognizing the unique (ormat. Most ether
clinicians wilt miss the child's cues and lens the opportunity for ready exploration of the traumatic mat2rial.
There have been occasions where we have tac• itly
colluded witti the child's avoidance and omitted soms
component of the interview, only to team afterward that
incomplete masterg of the trauma•related ansiety had
been achieved. We caution child therapists who teel
inclined to avoid the traumatic material to be aware of
countertransference identification witti
the affected child.
Particular attention must be paid to the tinel phase )f
the interview as incomplete closure threatens the
effectiveness of the session. Without proper closure the
child will be left struggling witti traumatic material
without adequate enhancement of ego Eenellen witti
which to bind his anxiety. In that situatien, the interview can be experienced as unsettling.
This technique hes been readily learned by a number
of child therapists and applied to their werk witti a variety
of traumatized children. We have noted that neme of our
tolleagues have voleed concern that the interview's focus
on the traumatic event could unduly upset an already
victimized child. Our extensive experience, supported by
the werk of several ether investigators ~ Ayalon, 1983;
Frederick, 1985) confirms that open discussion of the
trauma offers immediate relief and net further distrens
to the child. Recent adult investigators have suggested
that there mag be an optimum time to provide intervention
bevond which it is more difficult to achieve ego
restoration and improved affective tolerante ( Horowitz
and Kaltreider, 1979; Van Der Kolk and Ducey, 1984).
Chcldhood Trauma
because of its limited scope.
A nalogees .Model
In searching for an analogous model of therapeutic
trauma consultation, we have been impressed witti the
resemblance hefwem our technique and that adopted by
military psychiatrista for the treatment of eelthere who
have viswed a buddy trilled or maimed in combat (Fox,
1974; Glans, 1947, 1954; Grinker and Spiegel, 1943;
Hendin et al., 1981; Kardiner, 1941; Kolb and
Mutaiipassi,1982; Lifton,1982; Smith, 1982; Teicher,
1953: U.S. Public Health Service, 1943). The principles
we share include: 1) the witnessing of extreme violente
constitutes a unique, severe, psychic trauma; 2) the
importante
of
"front•line
intervention"
hefere
maladaptive ego resolution is organized; 3) einee "ego
contraction" is a primary consequente of the trauma,
major efforts are directed at ego restitution and toplog
enhancement; 4) mastery required an affectively experienced "reliving," including a comprehensive review
of the traumatic event; 5) aggressive themes, especially
of retaliation, threaten ego restoration and must be fully
explored; 6) an "outraged superego" responds to the
passivity of the trauma experience by insisting that more
should have been dove to save the victim, and its
demands must be relived by the interview; 7) absent or
pathologie grief results trom efforts
to
avoid re-evokine trauma ic' anxietv', and 8) sus. .
________________________________
tained mastery can only be achieved through reinte
ration info the groep, family, or eommunity.
Similar interview methode fellow Erom these printiples. We remaio heiibis: at timer insisting the child
continue, at ether timer participating theatrically in
the reliving of the episode, and occasionally
asneming the cols of an interestel bystander.
Whenever stifled grief emerges, we offer open
consolation and physical comfort. We underscore
the realistic teers associated witti a langer, and
normalize the aniiety in reexperienting the event in
session. However, the two techniques differ in
handling the opening phase of the interview. Since
eelthere display traumatic amnesie or disavowal,
narcosynthesis or hypnosis veere often net•
Recent reviews have discuseed the significant psychologica) impact for children of a velde variety of
traumatic experiences (Eth and Pynoos, 1985; Ten,
1984). There is tetter recognitien of the intrapsychic,
behavioral and physiologic changes that can otter.
There mag be prominent intrusive and avoidant phenomenon, increased stater of arousal and incident speeltic
new behaviors. Reporta of longterm effecta have included
pessimistic life attitudes, alterations in personality,
diminished self-esteem and disturbances in interpersonal
relationsbips. Because these treematic sequelae do net
necessarily pass witti time, there is an obvious need to
develop effective methode of early intervention.
The landmark paper on the treatment of trauma in
essary adjuncta to penetrating these lefmees. On the
ether hand, we have found that children remaio consciously aware of the traumatic event, though
'nishing its pain through the ure of lens)-in•fantasy or
,
avoidance.