Cross-examination of Child Witnesses
Joel Erik Thompson*
ABSTRACT: The cross-examination of children is difficult and
challenging. Differences between child and adult witnesses are
discussed and suggestions are made for the effective cross-exanimation
of a child in cases of alleged sexual abuse.
The cross-examination of a child, especially a child alleged to have
been the victim of a crime, and most especially when the child is
alleged to have been the victim of a sexual crime, is probably the most
challenging courtroom situation a criminal defense lawyer faces.
How does one cross-examine a child witness? At the risk of seeming
to trivialize the issue, I can only say that one does that the way
porcupines mate. Very cautiously.
As challenging as it has always been to cross-examine children,
especially the child victim, in Arizona it is now much more difficult
than it was prior to passage of the Victims' Rights Amendment. It
is now virtually impossible for a cross-examiner not to violate the
first rule of cross-examination: "Don't ask a question unless you
know the answer you will receive." Without pretrial
interviews, it is virtually impossible to anticipate answers to
significant questions. On the other hand, if it is any consolation
to defense lawyers, despite the opportunity to conduct pretrial
interviews and "practices" with these witnesses, prosecutors
often are as uncertain of the answer that will be forthcoming to their
Why is there uncertainty about what to expect from a child
witness? The problem is probably 10% the child, and 90% the result
of the examiner's technique and approach. Adults, in general, are
often not good at listening to children or communicating with them on
their level. We are much more skilled at dominating them,
lecturing them and interrogating them. Too often, children are
seen as different from adults in irrelevant ways. In truth,
children are simply shorter, younger and less experienced adults in most
All of the adult discomfort in communicating with children is
magnified immensely when the child is the alleged victim of a sexual
crime. Whatever basic discomfort existed is now compounded by the
natural empathy adults have for the need to protect these vulnerable
little folk from bigger people who hurt them. Add that to the
almost universal discomfort of talking about sexual topics in the
presence of strangers and you have a worst case scenario for most
Differences Between Child and Adult Witnesses
Children are more intuitive than adults. They are also more
attentive to nonverbal cues. As adults become more educated, sadly
they become more cerebral. We are often amazed by the perceptive
skills of the blind and assume they have been blessed with heightened
perception with their other senses to compensate for their loss.
Actually, they are merely not distracted by the purely visual and
continue to use all of their senses. Children are analogous.
Because their verbal skills are still in the early stages of
development, they have not learned to rely so much on what is said and
tend to be better tuned to how things are said. Voice tone and
inflection are much more significant in communicating with a child than
they are with another adult.
It is also obvious that it is much easier to confuse children with
big words because their vocabularies are often limited. There is
much discussion among experts of the "accommodation syndrome"
in children. This, in simplistic terms, is nothing more than a
child's wish to be liked by people in positions of authority (virtually
every adult to a child). As a result, children tend to say what
they think an adult wants to hear, regardless of the truth or accuracy
of their statement. It is really nothing significantly different
from an adult's response to questions from the boss or a police officer
after a traffic stop. It is simply a heightened observation with a
child because of the child's sharper intuitive skills.
A debate has long ranged about whether children
"lie." Sadly, there are still forensic experts who
testify that "children never lie." Such a position
demonstrates "a lack of neutrality or freedom from bias in workers
assessing sexual abuse complaints" (Quinn, 1988). One study
showed that in a non-criminal context (domestic relations proceedings),
the court counselors believed that in about 20% of the cases, sexual
abuse did not occur (Thoennes, & Pearson, 1988).
Whether children lie is probably more a semantic debate than a real
difference of opinion between professionals. I would submit that
children do lie, just as adults do lie. Children are simply less
sophisticated, and hence lie for somewhat different reasons. Any
parent who has not seen his children do something naughty and then
bold-faced lie that they didn't do it is either a candidate for the
Parents' Hall of Fame, or is immensely unperceptive.
Perhaps the major distinction that can be drawn is that children
often have little perception of the magnitude of the consequence of a
lie regarding molestation or related interrogation. They also
often lack the clear distinction between what they have personally
observed or experienced and what they have been told. An obvious
example is the three-year-old who will adamantly insist that there is a
Santa Claus, because they have been told that he exists by people the
child trusts and the child has seen persuasive evidence. Thus,
often children make statements that are not factually accurate, but they
are not "lies" because the child lacks the intention to
willfully mislead or deceive.
Children may also be unable to perceive the difference between an
adult who intends to be truthful but misspeaks and thus says something
inaccurate and the adult who intentionally misstates the truth.
They may call both lying because they fail to distinguish the
significance of the adult's intention.
Similarly, the common prosecution chestnut of showing that a child
witness knows the difference between the truth and a lie is really of no
use in evaluating the credibility of the witness or the child's
competency to testify. A child may truly understand the difference
between the truth and a lie, yet have no comprehension of whether what
an adult says to them is one, the other or both.
In the end, honesty is not an age-dependent characteristic. We
do not grow more honest as we grow older, in fact we may grow less
so. To any given question, either an adult or a child who is
honest may respond with a lie, for a variety of reasons.
Conversely, a dishonest child or adult may give an honest answer to the
same question for very different reasons. Honest people don't
always tell the truth and dishonest people don't always lie.
It is often asserted that children are telling the truth about sexual
happenings alleged to have occurred to them or in their presence because
they lack the sophistication and knowledge to make up the
incidents. Again, this is a naive and rather simplistic
presumption. Perhaps it is true for a very young child, assuming
they do not share a bedroom with sexually active roommates, have no
older siblings or playmates and never are up after 9:00 p.m. in a
household with access to cable TV. For good or ill, part of the
earlier sophistication and education of our children is an awareness of
a great deal of sexual fact at surprisingly early ages. Children
are as bombarded with sex and sensuality in advertising, on TV and in G
and PG-13 rated movies as any adult.
One of the true differences between young children and older children
or adults is the ability to distinguish what is relevant from what is
irrelevant. Thus, an older child may give a fairly concise
description of being molested. By contrast, a young child could be
expected to give a variety of irrelevant factual information. An
example would be a young child testifying about a molestation intermixed
with observations about the child's puppy, statements about flowers in
the garden or other events or observations that have no factual or
temporal connection to the molestation.
In cross-examining a child witness, it is critical to remember that
the jury will not tolerate the examiner attacking the witness.
There are often witnesses that invite their own destruction when they
testify so much so that when it happens the jury silently
applauds the occurrence. This does not happen with child
witnesses. The jury does not identify with lawyers.
Americans have a tradition of rooting for the underdog. In a
battle of wits and courtroom skills between a lawyer and a child, the
lawyer is assured of never being viewed as the underdog.
Further, and perhaps most significantly, even if the examiner succeeds
in devastating the credibility of what a child witness has testified as
occurring, rest assured that the jury will have no difficulty creating a
plausible explanation for the deficiencies in the child's
testimony. The ultimate goal in cross-examination of a child is to
establish a credible contradiction in fact. If the child testifies
that he knows it was Thursday night that the events occurred because
that is the day his favorite TV show is on and he was watching the show,
testimony from the station manager that the show was not broadcast that
night will accomplish far more than attacking the child and his poor
Communication Errors When Talking With Children
Adults must be aware that time is a very difficult concept for
children to comprehend. Any parent traveling any distance with a
child can attest to the fact that a child who is told "we will
arrive there in four hours" will ask again in 20 minutes, "are
we there yet?" To a child, especially a very young child, the
term "o'clock" is as apt to refer to the Irish family next
door as it is to time. Thus, if a child gives you a non-childlike
answer, such as, "it was Thursday at 3:00 p.m.," a
cross-examiner should immediately suspect adult input.
Similarly, children have great difficulty recalling events in
sequence. This becomes further exacerbated when a child has been
abused in repetitious events, especially by a single abuser known well
to the child. The average length of an abusive relationship for girls
is 31 weeks and the abuser is an individual well known to the child in
75-80% of the cases (Finkelhor, 1979; Mian, Wehrspann, Klajner-Diamond
& LeBaron, & Winder, 1986; Tsai & Wagner, 1978).
Once again, if the child seems uncharacteristically able to put
events in a sequence, suspect adult interference with the child's
natural recollection process. Just as some lawyers seem unable to
understand that even fairly educated people are confused about whether
subsequent means before or after, many more forget that to a child even
the concept of before and after are muddled in the imprecision of a
child's concept of time. Much more will be gained by trying to ask
about events that could have happened at the same time when talking with
children. "Did this happen at the same time Daddy was yelling at
the puppy?" will elicit a much more usable factual response than
asking, "Did this happen before or after breakfast?"
Children also have difficulty with numbers. A young child may
be able to count to 10, but if asked how many times he was molested, the
same child is totally incapable of giving an answer. If tested
further and asked to hand the examiner 5 of the 10 pennies placed before
him, the child would be immobilized.
Family relationships are also confusing to children. Daddy's
mommy. Is that the child's grandmother, or Daddy's wife? To
a child, a mommy is the woman living in the house who is responsible for
domestic matters like meals, baths and laundry. If questions are
not formulated with great care, it takes very few questions to create a
total mishmash out of the child's description of events and
Bear in mind also that the worldly perspective of a child is very
different from an adult's. Adults see the world from 60 to 72
inches above the floor. Children see the world from 30 to 36
inches above the floor. Where visual observations are critical,
part of your pretrial investigation should be "standing" in
the child's position while on your knees.
Areas where children are naturally inaccurate in their testimony are
often more persuasive of the truth of the allegation than the
opposite. For example, a three-year-old who says "then the
man peed," rather than he ejaculated, is more in line with a
child's perception of the function of a penis. Again, if the child
seems unnaturally accurate regarding such testimony, the careful
cross-examiner should become suspicious of adult contamination of the testimony.
Questioning Techniques for Child Witnesses
It is essential to anyone questioning a child witness to bear in mind
that children are hostile to all questioning. Questions to
a child are often a preamble to criticism, orders, complaint or
judgmental observation by parents, teachers or peers. Thus, a
child may give an evasive response to a question an adult would consider
a simple request. For example, in response to the question,
"What are you doing?" a child is likely to give an obviously
inaccurate, but non-committal answer of "nothing." Such
an answer insulates the responding child from the myriad of negative
trailers the inquisitor may have in mind.
Suggestive questions also cause problems even in children who are
accurately recalling events and honestly trying to report them. A
textbook example is the experimentation of Varendonck, an expert called
to a Belgian courtroom in 1910 to evaluate the credibility of child
witnesses in a murder trial. Varendonck asked 18 seven-year-olds
the color of the beard of their teacher. Sixteen responded
"black" and two did not answer. The teacher was clean
shaven. The suggestion that the teacher had a beard caused these
children to respond with the most likely answer, rather than the
accurate one because their memory was challenged by the question.
Since the children lacked the confidence to attack the question, they
accepted it and simply tried to answer it as though the question were
Another area to be aware of when questioning children is the
difference in moral code or values between adults and children.
For example, adults will frequently espouse honesty while routinely
stealing inexpensive items (or worse) from the office. To a child,
the dichotomy between the statements and the action is confusing.
To an adult, the rationalization is made that bringing pencils home from
the office is acceptable because office work is often done in the
evenings at home.
Perhaps a more illuminating example concerns the "Don't drink
and drive" campaign. An adult knows that drinking in this
context refers to alcoholic beverages. But a three-year-old will
tell Daddy very authoritatively that he should not be drinking from his
Coke can while driving. Try explaining the distinction to that
child and expect a look of serious skepticism.
Cross-Examination Goals and Techniques with Child Witnesses
We have established that attacking child witnesses is
fruitless. They can be neutralized, as any other witness, with
independent evidence establishing that their testimony is impossible or
unlikely. With skillful attention to well posed and carefully
presented questions, a child can be shown not to have credible evidence
to present (assuming that is the case). Sometimes the same can be
established even when it is not true. But, it is equally possible
to present a lying witness in a manner that makes the testimony
Many lawyers presume that Arizona Revised Statutes §-4061 with its
absolute assertion that "In any criminal trial every person is
competent to be a witness" prevents exclusion of a child
witness. Should you be faced with a judge who believes that no
inquiry as to the competency of any witness may be undertaken, let me
suggest that you propose to present the testimony of your phrenologist
who will testify that after examining the bumps on the skull of the
child victim, your expert witness has determined the child is
lying. Once the judge has ruled that your expert is not competent
to testify, and you have established that the language of the statute
means only "every person with relevant evidence to present" is
competent to testify, you may then approach whether the child witness
has relevant testimony to present.
Rule 601 of the Arizona Rules of Evidence clearly states that
"Every person is competent to be a witness except as otherwise
provided in these rules or by statute." Clearly, the statute
referred to has not changed the rules of evidence at all. Rule 602
of the Arizona Rules of Evidence states "A witness may not
testify to a matter unless evidence is introduced sufficient to support
a finding that the witness has personal knowledge of the matter."
A child witness, no differently from an adult witness, is not
competent to testify unless he has personal knowledge of the
events. If a child's testimony lacks support to show that it is
based on personal knowledge, it is as incompetent as any adult's
testimony lacking similar support.
To effectively cross-examine a child, use simple language.
Learn from the advertisers and newspapers. Ask short, direct
questions. Anticipate that children may be stymied by the
questions "who," "what," "where,"
"when," and "why." Good child witness
questions are five-word sentences. As the number of words in the
question rises, the ability of the child to understand the question
Avoid confrontation with the child. It is OK to be an
adversary, it is damaging to be hostile. Watch the tone of your
voice. Children intimidate very easily, especially if you are male
and have a deep voice. (Children are usually closer to mothers
than fathers, and hence find motherly people less threatening when they
Avoid dominating the child on the witness stand. Even a short
adult towers over a small child, even when the child is seated in the
witness chair on pillows and telephone books. If you must give the
child a document, consider having the judge or court staff actually hand
it to the child.
Listen to what a child says, and try to understand what the child means.
Often children say things that they understand, but that adults give
adult meaning to. (A child who tells her teacher she did not get
her homework done because she had to sleep with Daddy will capture the
teacher's undivided attention.) Upon questioning the child
(objectively and neutrally), the teacher may find that the family lives
in a one-room studio apartment and everybody sleeps in the same
bed. The teacher may also find out that absolutely no impropriety
occurred and her student was simply wishing she had her own room.
Assumptions by adults create disastrous consequences whether it is by a
teacher, a social worker, a police officer, a prosecutor or a
cross-examining defense lawyer. Listening can tell you much
more than assuming. Hear beyond the words and grasp the
child witness's meaning.
Do not ramble in questioning children! Have a plan and
purpose to your questioning. If you don't need to ask the
question, don't ask it! Giving children the opportunity to
add additional information as they answer a pointless question will
seldom produce a beneficial result, Adopt an inquisitive,
interested and conversational demeanor and voice tone when questioning
children. If they don't want to talk to you, even being told by a
judge to answer your question will only produce sustained silence.
When stressed, most children know that silence is safety. And you
know you are dying on your feet.
Finkelhor, D. (1979). Risk factors in the sexual victimization of
Child Abuse & Neglect, 4, 265-273.
Mian, M.. Wehrspann, W., Klajner-Diamond, H., LeBaron, D. &
Winder, C. (1986). Review of 125 children 6 years of age and under who
were sexually abused.
Child Abuse & Neglect, 10, 223-229.
Quinn, K. M. (1988). The credibility of children's allegations of
sexual abuse, Behavioral
the Law, 6, 181-199.
Thoennes, N., & Pearson, 1(1988). A difficult dilemma: Responding
to sexual abuse allegations in custody and visitation disputes. In D. J.
Besharov (Ed.), Protecting Children from Abuse and Neglect: Policy
and Practice ()
(pp. 91-112). Springfield, IL: Charles C. Thomas.
Tsai, M., & Wagner, N. (1978). Therapy groups for women sexually
molested as children. Archives of Sexual Behavior,