Expert Witness: Challenging the forensic interview

By Michael G. Brock

“There is one thing a professor can be absolutely certain of: almost every student entering the university believes, or says he believes, that truth is relative. If this belief is put to the test, one can count on the students’ reaction: they will be uncomprehending. That anyone should regard the proposition as not self-evident astonishes them, as though he were calling into question 2 + 2=4. These are things you don’t think about. The students’ backgrounds are as various as America can provide. Some are religious, some atheists; some are to the Left, some to the Right; some intend to be scientists, some humanists or professionals or businessmen; some are poor, some rich. They are unified only in their relativism and in their allegiance to equality. And the two are related in a moral intention. The relativity of truth is not a theoretical insight but a moral postulate, the condition of a free society, or so they see it.”

—Alan Bloom
Introduction: The Closing of the American Mind

“An ideologue—one who thinks ideologically—can’t lose. He can’t lose because his answer, his interpretation have been determined in advance of the particular experience or observation. They are derived from the ideology, and are not subject to the facts. There is no possible argument, observation or experiment that could disprove a firm ideological belief for the very simple reason that an ideologue will not accept any argument, observation or experiment as constituting disproof.”

—James Burnham

The Suicide of the West:
An Essay on the Meaning and Destiny of Liberalism (p. 108)

“The courtroom oath—to tell the truth, the whole truth and nothing but the truth—is applicable only to witnesses... because the American justice system is built on a foundation of not telling the whole entire truth.”
—Alan Dershowitz,

The Best Defense (New York: Vintage), 1983-5-12, p. xiv.

“As one civil-liberties lawyer, who is concerned about the sometimes vigilante attitude toward accused rapists, puts it: ‘Some people regard rape as so heinous an offense that they would not even regard innocence as a defense.’”
—Alan Dershowitz

New Dangers Are Evident in Rape-Case ‘Reforms’ published 1985-04-08 Los Angeles Times

“The goal of a forensic interview is to obtain a statement from a child, in a developmentally-sensitive, unbiased, and truth-seeking manner, that will support accurate and fair decision-making in the criminal justice and child welfare systems.”

—Introduction

Michigan Forensic Interviewing Protocol


I used to send my Criminal Sexual Conduct Cases to Terry Campbell, who was well qualified, good at it, and liked doing them.  When he died I looked around, but wasn’t really confident of the people who were taking these cases, or satisfied with the job they were doing.  So, recently I began taking CSC cases again, and researching/catching up on the forensic literature.  There are some changes in the process, though I don’t know that these changes are all improvements.  Some of the fine tuning in protocols mentioned by Debra Poole in her recent book1 indicated advancements that would undoubtedly be helpful for someone who had the basics of interviewing and wanted to refine the process.  Dr. Poole also maintains an objective/neutral position with regard to the outcome of the interview, the importance of the child’s narrative, exploring alternative hypotheses, and (most importantly) avoiding leading questions.

Some of the newer protocols, apparently in the interest of being more prosecutor friendly (the main consumers of their product), do not hold to these basic scientific principles.  Michael Lamb’s NICHD Protocol2, for example, which is in wide use, offers no techniques for exploring alternative hypotheses, and allows what I consider extremely leading, even coercive questions.  Whereas Dr. Poole advises not to lead children by questioning them about issues they have not yet raised, Lamb’s Protocol specifically provides for it:

“– Eliciting Information that has not been mentioned by the child. You should ask these focused questions only if you have already tried other approaches and you realize that some forensically important information is still missing.   It is very important to pair option-posing questions with open invitations (‘Tell me all about that’) whenever possible3.”

Note that Lamb is stating here that it is OK to seek information that has not been mentioned by the child.  This is clearly hypothesis confirming rather than hypothesis exploring behavior.  When he says that forensically important information is missing, he means information that has been provided to him by other sources that the child has not yet confirmed.  If he were exploring alternative hypotheses, he might consider that the information provided to him by other sources was either contrived by that source, or obtained because of hysterical and extremely leading questions, which is frequently the case with parents who are coaching their children in acrimonious custody disputes, or questioning children they suspect of having been abused.

As it is, telling a child that someone else—an authority figure that the child may feel guilty about contradicting—has told the interviewer a different story, is very coercive and apt to support a false allegation.  This is particularly true if the child has been coached and/or pressured by a parent or family member to provide a certain response.  Keep in mind also that the persons conducting the interview, and those deciding whether the interview is good “science” are members of the prosecution team, who are not looking for exculpatory evidence, but have a confirmation bias.  It doesn’t help the cause of justice if the interviewing protocol is also biased.

Since I have been doing these cases again, it continues to surprise me that so many defense attorneys know so little about interviewing protocol.  I talked with forensic psychologist James Bow about this yesterday on the phone and he agreed that this was shocking to him.  Possible explanations he put forward are:

• Everyone drinks the Kool Aid—pro prosecution propaganda is ubiquitous and many defense attorneys may question their client’s innocence of even ridiculous charges because they may consciously or unconsciously believe that everyone is guilty.

• They are intimidated by science, have little feel for it, and believe that if they approach it at all, it is better left to the expert.

• They have always believed that forensic science is the domain of the prosecution, and that it is more likely to be useful to their adversary than to themselves.

• They equate challenging the forensic interview with attacking the child, which is rarely if ever beneficial to the defense.

• They believe the less time spent on the prosecution’s allegations, the less those allegations are reinforced in the jury’s minds.  Don’t keep reminding them.  Above all, don’t play the entire forensic interview video (if one exists), or offer the prosecution a chance to do so.

I have heard this concerns expressed by defense attorneys and am reluctant to question them.  I’m not a lawyer.  Moreover, I don’t want the defense attorney to blame me if I say I think he should put me on the stand, play the video, challenge the interview, and he winds up losing the case.  On the other hand, I don’t know how many of these trials are actually won by the defense.  It has also been my experience that when I testify the jury is out longer and produced more favorable verdicts that in cases where I’ve been retained and not testified, and that the jurors at least have a competing theory to consider.  I’ve read articles by trial attorneys who express this understanding of the jury’s mindset:

“All too often, criminal defense attorneys rest their case and await a favorable verdict—confident that they have rendered a critical state witness unbelievable—only to lose. Why? Unfortunately for these defense attorneys, undermining the credibility of a critical witness is not sufficient to prevail. True, the prosecutor has the burden of proof, but both parties have a more important burden: a good story. Parties must tear down, but they must also create. In the absence of an alternative explanation presented on behalf of the defendant, the jury gravitates toward the only story, the State’s. A persuasive counter-narrative has essential elements. It provides a compelling explanation for what happened. It has factual integrity. And it supports cognizable defenses: Someone else did it, the defendant has an alibi or an affirmative defense, the police conducted a flawed investigation, or no crime actually occurred. However innocent or wrongfully charged, a defendant without a counter-narrative will lose.”4

OK, but how does one challenge the forensic interview?  I have read a large enough chunk of the forensic literature to realize that for the most part, everyone is on or near the same page with regard to what constitutes best practices, and even what the forensic interviewer ought not to do.  However, there is very little available that specifically tells someone trying to evaluate a forensic interview what constitutes a bad interview.  And keep in mind, this is the defense attorney’s burden.

Most prosecutors and their teammates—including forensic interviewers—have never heard a false allegation, and have never seen or done a bad interview.  If defense counsel does not point it out to the jury, no one will.  Nor can you expect to get any help from judges, who may be skeptical of the evidence, but still have to get elected, and many of whom also drink the Kool Aid of 24/7 pro-extreme-Right-and-extreme-Left-wing-law-and-order feminist-propaganda.  This is the only respect in which our otherwise hopelessly divided country is united, we all absorb, to a greater or lesser extent the mass hysteria that will likely destroy our nation and Western Civilization sooner rather than later.5

For those of you who really don’t think our society is that crazy, remember the words of Martin Niemoller, an early supporter of Hitler’s:

“First they came for the Socialists, and I did not speak out—

“Because I was not a Socialist.

“Then they came for the Trade Unionists, and I did not speak out—

“Because I was not a Trade Unionist.

“Then they came for the Jews, and I did not speak out—

“Because I was not a Jew.

“Then they came for me—and there was no one left to speak for me.

The above statements notwithstanding, there is some literature about challenging the forensic interview, and the testimony of children in general, both from lawyers and mental health professionals, that bears reviewing and that could be helpful in seeking the truth of whether or not a child’s evidence can be relied upon.  This is not to say that a child has to be lying, although they will lie, and about large things as well as small.  Historically, in fact, children’s testimony has been considered unreliable in court.6  What is supposed to make current cases different is the so called science of the way we obtain their information.  However, science rarely comes in to play in my experience, whereas a host of other things do: contested custody, a child’s embarrassment at being caught in sexual play (which, of course, none of us ever did when we were children), hysterical parents who see a child molester behind every bush and who badger their children with merciless persistence until they reveal what the parents most fear to hear (at which point they are both satisfied and destroyed), and, in one instance that I still recall from my days in family court, the promise of anything the child wanted from McDonald’s if she told me that her mother had touched her inappropriately (The PS worker later called me to relay this info).  They can be bribed, coerced, brainwashed and threatened, make up stories for attention or to get out of trouble, be subjected to counselors who practice forensic in the name of therapy, and ultimately reach a point where they have no idea what the truth is because they have had memories implanted that have now become “real memories,” but these possibilities are never explored by the forensic interviewers.

I had one case recently where the child told the forensic interviewer that her mother had told her what to say.  The forensic interviewer, a police officer, quickly assured her that she wasn’t in any trouble, nor was her mother.  That the police officer was to keep her safe, and then she quickly steered the child back to the target of the investigation, her father.  When she could not get the allegation she wanted from increasingly coercive interviewing, she provided a large toy animal for the child to confide her secret to, and absolute violation of proper forensic interviewing protocol, which warns interviewers to stay away from anything smacking of play or fantasy:

“Improper Interviewing: As we use the term, improper interviewing techniques consist mainly of things that interviewers should not do. Specifically, certain techniques have been widely criticized because they have the potential, either real or perceived, to elicit false allegations from children (Ceci & Bruck, 1998, 1995; Everson, 1997; Myers, 1996; Myers, Saywitz, & Goodman, 1996; Reed, 1996). Garven, Wood, Malpass, and Shaw (1998) have suggested that most of these techniques fall into four broad categories: Suggestiveness, influence, reinforcement, and removal from direct experience (SIRR)

“Suggestiveness: This occurs when the interviewer; rather than the child, introduces new information about the topic of concern into an interview. Fox example, “Did he touch you on the bottom?” would be a highly suggestive question in a sexual abuse interview if the child had not already mentioned inappropriate touching. Many studies have shown that interviewer suggestiveness can reduce the accuracy of children (Cassel, Roebers, & Bjorklund, 1996; Poole & Lindsay, 1995; see summaries by Ceci & Bruck, 1993, 1995; Poole & Lamb, 1998). One or two mildly suggestive questions may be necessary to get the ball rolling with some children or focus their attention on a particular abuse-related topic (Home Office, 1992; Jones, 1992; Warren, Woodall, Hunt, & Perry, 1996; Wood, McClure, & Birch, 1996). However, suggestiveness can seriously compromise accuracy when the child is 4 years old or younger; or has been interviewed repeatedly (Ceci & Bruck, 1993, 1995; Garven et: at, 1998; Myers et al., 1996).
“Influence. This category encompasses various techniques that place undue social pressure on a child during an interview. These problematic techniques involve social influence or persuasion (Brewer & Crano 1994; Cialdini, 1993) and include (a) inducing social conformity by telling the child what other people believe or have said about the topic of concern (Garven et al., 1998; Myers, 1996, p.. 218), (b) eliciting obedience to authority by telling the child the interviewer’s point of view, and (c) inducing stereotypes by describing an alleged perpetrator in negative terms to the child (Leichtman & Ceci, 1995).

“Reinforcement. As has long been known, reinforcement in the form of tangible, promised, or implied punishment or reward can have a powerful influence on behavior (Ettinger; Crooks, & Stein, 1994). More specifically, recent research has shown that reinforcement can elicit false allegations of wrongdoing from children during interviews (Garven, Wood, & Malpass, in press; Garven et al., 1998). Several improper interviewing techniques constitute forms of reinforcement (see also Lamb, Steinberg, & Esplin, 1995; Myers, 1996, p 218): (a) praising the child for making allegations; (b) implying that the child can demonstrate helpfulness or intelligence by making allegations; (c) criticizing the child’s statements or suggesting that they are false, inaccurate, or otherwise inadequate; (d) giving tangible rewards (e.g., stickers or food) to reward disclosure; (e) limiting the child’s mobility (e.g., delaying a visit to the bathroom or return to home) until he or she has discussed issues of interest to the interviewer; (subjecting the child to physically or verbally stressful stimuli during the interview (e.g.. calling the child a liar); and (g) repeating a question in a way that suggests the child’s first answer was incorrect or otherwise unsatisfactory (Cassel et al., 1996; Poole & White, 1991, 1993; Siegal, Waters, & Dinwiddy, 1988).

“Removal from direct experience. In forensic interviews with both adults and children, the witness is usually asked to give a direct description of what he or she has observed and experienced. However; some child interviews deviate from this direct format and instead take an oblique or indirect approach to information gathering. Problematic indirect interviewing techniques include (a) inviting the child to speculate about what might have happened, rather than to describe what the child actually observed; (b) encouraging the child to pretend or engage in imaginative play as part of the investigative interview; and (c) interviewing the child using puppets (but see Jones, 1992, P. 40)”7

Note that the interviewer violated Michigan Protocol by failing to even consider the possibility that the child had been coached, even after the child’s statements to that effect, and by steering he interview in the direction she wanted it to go; that is, toward confirming the only hypothesis she was willing to consider, that her father had molested her.  She also violated best practices by blotting out the child’s visibility and audibility with the oversized toy, also bribing the child to provide information she had not been willing to provide, and then encouraging the child to engage in fantasy instead of the dead serious reality that should have been the focus of this interview.

Michigan Protocol author Debra Poole has never forgotten the lessons of her past research, nor ignored the fact that she is a scientist and not a member of prosecution team.  She has long recognized that children provide misinformation for a variety of reasons:

“Ceci and Bruck (1995), for example, compared descriptions of coding categories in studies of false allegation rates with the studies’ final conclusions. In some cases, conclusions about false allegations reflected only cases in which there was intentional lying, with authors dismissing cases in which false allegations were made for other reasons. When these other mechanisms were included (e.g., honest misunderstandings of children’s statements or problems caused by reliance on presumed indicators of abuse followed by suggestive interviewing), rates of false allegations rose considerably (e.g., from 6% to 23% in Jones & McGraw, 1987; from 8.8% to 35% in Faller, 1991).” (P. 2, 3)

“In recent years, however, there have been a number of compelling demonstrations that children do sometimes falsely report events that involve bodily touch or harm, that such reports are alarmingly easy to produce—at least at young ages—and that these reports often contain many of the characteristics considered to be ‘diagnostic’ of true reports.” (P. 3)

“Unhappily, many experts who testify about the characteristics of abused children claim to do so on the basis of their ‘experience,’ but claim to have little or no experience with children who weren’t abused but who at one point claimed to have been. Hence, such experience is—not just ‘strictly speaking’ but profoundly—irrelevant to ‘the determination of an action.’ Ironically, such pseudo-experts often try to support their alleged expertise by claiming that almost all the children they see claiming abuse have in fact been abused. If taken seriously, such a claim should automatically disqualify them as having any experienced-based expertise in the matter of most urgent interest to the court, which is the rational determination whether or not the child has been abused as claimed. (pp. 6-7)” (P. 21)

“[t]his research has demonstrated how the accuracy of children’s reports can be dramatically shifted by adult influences. Adults alter the a priori probability of abuse in investigated cases by their referral strategies, the content and characteristics of children’s reports by their interviewing strategies, and the outcome of individual lives by their tolerance for false-positive versus false-negative errors. It is important to note that there is no reason to assume that efforts to reduce the probability of false positives will necessarily increase the probability of false negatives. What is required is that assessors do not merely alter the stringency of their response criterion, but rather that they qualitatively change and improve the procedures by which assessments are made. Pursuit of this goal is clearly in the best interest of both abused and non-abused children and of society at large.” (P. 22)8

It might be expected that there would be a great deal of interviewer bias, given that the determination of what constitutes good science is the domain of the prosecutor’s office.  Prosecutors are neither scientists, nor impartial.  Yet the same well educated attorneys who would be outraged by Tom Brady deciding how much air should be in a football in a championship game, are willing to quietly and complacently acquiesce to the prosecutor’s office about a matter of far more importance: the fairness of the judicial system.  Research shows that defense lawyers rarely delve into the process by which forensic evidence is obtained from a child, nor do they employ the techniques that would give them access to reliable evidence from child witnesses in court.  This is especially surprising given that there are so many interviewing errors, and that evidence of conscious or unconscious bias is so apparent in the majority of forensic interviews I have observed.  Note the following study of attorney cross examination practices9:

“What was striking was that both attorneys [prosecution and defense] predominantly asked questions that simply asked for a “yes” or a “no” (two thirds of prosecution questions and over 80% of defense questions), and children typically provided unelaborated answers. The attorneys’ emphasis on yes–no questions meant that they [the attorney] were responsible for generating details of the interactions. Furthermore, children virtually never spontaneously referred to conversations (less than 1% of the time), and thus were dependent upon the attorneys’ questions to do so. The emphasis on yes–no questions probably decreased both the productivity and accuracy of responses, because recognition questions produce fewer details and less accurate details than recall questions (Lamb, Hershkowitz, Orbach, & Esplin, 2008). Although attorneys were less likely to ask older children yes–no questions, they nevertheless did so 41% of the time, and there were no age differences in the proportion of other questions asked, including wh- questions [who, what, where, how, when, etc.] tapping recall memory.”

All forensic research emphasizes how much more information—and more accurate information—can be obtained from open-ended questions than from short-answer, one word, or yes or no questions.  If an attorney is looking for a way to get at the truth of a child’s report without appearing to be beating up on the child, this is a very good way to do it, but attorneys appear to have very little faith in science, even though answers from scientific questioning are the most reliable evidence a child can produce.

In my experience, a child who is telling the truth of their own experience can provide a reasonably coherent narrative with some detail unique to their case.  One way of observing this is to compare the child’s practice narrative (if one is elicited, as all protocols recommend) with the narrative of the alleged abuse.  In cases where the allegations are questionable, children cannot provide a credible narrative, they either repeat key phrases that they have been fed by someone (daddy touched my pee-pee, white stuff came out, he put his thing in my thing, etc.) without any context, or they provided a rambling narrative of possible truths mixed with obvious fantasy, but lacking any real linear progression.  The research tends to show that what makes an adult’s report believably and probable, is the same thing that makes a child’s report credible.

In two credible stories of abuse that were told to me, a younger child said an older boy came into his room at night, shined a flashlight on his face, and rubbed his penis on his face.  He also remembered that the older child had a pattern of little animals on his underwear.  Such realistic detail is seldom found in false reports of young children, and even older children’s fabrications are generally full of holes.  In another story told to me by a 12 year old girl, an older babysitter had told her to go upstairs and take her pants off.  Then he told her to hold onto the bed board, got behind her and rubbed his penis on her anus.  Again, there is linear progression, and credible detail.

What juries find credible from child witnesses was the subject of another study of child abuse cases in Scotland10:

“In a study by Cashmore and Trimboli (2006), jurors frequently justified their judgments regarding the truthfulness of children’s testimony on the basis of the consistency of details about dates, places, or clothing, and these judgements in turn predicted their verdicts. Another study by Connolly, Price, and Gordon (2009) found that, although inconsistencies were mentioned in 75% of the judicial comments regarding verdicts in historical child sexual abuse cases, and were twice as common when there were acquittals rather than convictions, complainants’ emotional behavior during the events and at the trial were more strongly associated with verdicts than the frequency of inconsistencies in witnesses’ statements. However, the mean age of witnesses at the time of the trial in Connolly et al.’s study was 25.93 years, which could indicate that adults were more consistent witnesses than children, or that juries place a higher emphasis on consistency when assessing the testimony of child witnesses. Therefore, the disproportionate number of self-contradictions defense lawyers refer to using credibility-challenging questions focused on peripheral details or repeated conversations could have strong negative effects on fact-finders’ perceptions of the truthfulness of children’s testimony and on trial outcomes.” (P. 21)

Besides the fact that jurors do care about consistency in children’s’ testimony, note two important factors addressed by the authors.  First, the child’s emotional state has a profound impact on the jury.  One lawyer said in an article I read in preparation for writing this paper that a crying child is “bullet proof.”  Indeed, who could not be moved by the specter of a child crying on the stand?  No one is that hard hearted and rational.  But some children are criers and lawyers have to find a way to deal with it.  One thing the defense lawyer has to keep in mind is that he is not attacking the child; she is a victim of someone.  But is she the victim of your client, or of malicious, or even well intentioned programing?

I gave testimony recently in a case with this type of child witness.  In the forensic interview, the child’s story was embellished beyond all reason by the interviewer; the story was literally constructed by the interviewer.  I was able to explore this with the jury, but there were equally important pieces of information that I was not allowed to go into.  The child said in the forensic interviewer she was raped daily, then twice in one day in the preliminary hearing, and only once when she got to the stand.  I wasn’t there for the child’s testimony, but the morphing of the child’s story over time is a sign that the child’s testimony lacked credibility.  Moreover, she had been questioned by an anxious relative, to whom she told another story.  How a lawyer deals with these inconsistencies delicately is certainly something he has to consider before trial.  It can’t be avoided.  Below are some suggestions from a law school professor that might be applicable11:

“Child abusers are not the only adults who pressure child witnesses to provide false testimony. It is unfortunately true that some parents embroiled in child custody litigation fabricate allegations of abuse and persuade children to accuse an innocent parent. The target of such an accusation has a right to uncover the truth by impeaching the child’s testimony with proof of the coercion which produced the inaccurate testimony.  Coercion may be established through examination of the child. Alternatively, counsel may admit extrinsic evidence to establish coercion. Since this form of impeachment relates directly to the credibility of the witness, the collateral fact rule does not apply, and extrinsic evidence may be admitted.  A majority of courts require counsel to lay a foundation before offering extrinsic evidence. The trial judge has discretion to limit extrinsic evidence used to impeach. In the case of coercion, however, rather broad latitude should be accorded the impeacher.

“Coached Testimony: A child whose testimony is the product of improper coaching is a biased witness in that her or his testimony reflects an advocate’s position rather than an objective statement of the facts. A party against whom coached testimony is offered has a right to examine the witness in an effort to disclose coaching. Furthermore, counsel may offer extrinsic evidence to establish coaching. The collateral fact rule should not apply to impeachment for coaching because coaching relates directly to the credibility of the witness. As is true with impeachment for bias, interest, and coercion, the court may require a foundation before counsel resorts to extrinsic evidence to establish coaching.” (P. 924-926)

“Rule 806 provides that when a hearsay statement is admitted in evidence, ‘the credibility of the declarant may be attacked, and if attacked may be supported, by any evidence which would be admissible for those purposes if declarant had testified as a witness.’ The impeaching attorney may employ any of the five modes of impeachment: Inconsistent statements, character, bias, defects in capacity, and contradiction. Rule 806 eliminates the foundation requirements which are often required for impeachment.” (P. 941)

“The suggestibility of children is important for three reasons. First, the cross-examiner can occasionally undermine the credibility of a child’s testimony by demonstrating that the child is highly suggestible. Second, some suggestible children can be led to alter their direct testimony through skillful use of suggestive questions during cross. Third, in some cases the cross-examiner’s pretrial investigation reveals that adults such as parents, investigators, or mental health professionals conducted highly suggestive interviews of a child; interviews which may have distorted or even obliterated the truth. In these cases the cross-examiner elicits information from the child about the number of interviews and what occurred each time. This testimony may be followed by cross-examination of the individuals who interviewed the child. Cross-examination of the adults is designed to elicit evidence of improperly suggestive interview techniques. (emphasis mine)

“As stated above, the cross-examiner can capitalize on the suggestibility of some children. Consider the following excerpt from the cross-examination by defense counsel of a young child at a preliminary hearing:

“Q: Do you remember the last time that we talked?

“A: Yes, I do.

“Q: And when I had that different colored suit on, do you remember that?

“A: Yes, I do. I really do.

“Q: The one that was red?

“A: Yes. I remember it.

“Q: You remember it?

“A: Yeah.

“Q: Do you remember the stripes I had on the other suit were red and they went around my legs that way? Do you remember that?

“A: Uh-huh (affirmative).

“Q: And do you remember we talked about going and getting ice cream?

“A: Yes. I remember that.

“Q: And that me and you were going to get an ice cream, and do you remember my friend that was with me? The other guy that had a beard like I do? Do you remember that man?

“A: Yes. I do remember that man, but I don’t remember his name.

“Q: And the three of us were going to go get an ice cream. Do you remember that?

“A: Yes. I remember that.

“Q: How come we didn’t go get the ice cream?

“A: I can’t remember.

“Q: Why didn’t we get the ice cream? What happened?

“A: I don’t know.

“Q: We just didn’t go, did we?

“A. No, we didn’t.”

“In fact, the attorney had never met the child. There was no red suit with horizontal stripes, no aborted trip to the ice