Chapter 7, excerpted from True and False Accusations of Child Sex-Abuse,
© 1992 Creative Therapeutics, Inc., Cresskill, NJ




During the last few years a conflict has been raging over the value and place of anatomically correct dolls (more recently referred to as anatomically detailed dolls) in child sex-abuse evaluations. Studies are coming in from both camps, each claiming that the findings support its position. The American Psychological Association (APA) has periodically issued position statements on the use of these dolls, but has yet to come up with a definitive statement. The latest proposal (February 8, 1991) urges ``continued research'' and hedges a final decision as to whether the dolls per se are ethical to use. The APA recommends that documentation be provided in association with the use of these dolls:

We recommend that psychologists document by videotape, (whenever possible), audiotape or in writing the procedures they use for each administration. Psychologists should be prepared to provide clinical and empirical rationale (i.e., published studies, clinical experience, etc.) for procedures employed and for interpretation of results derived from using anatomically detailed dolls.

Considering the weaknesses and criticisms that can justifiably be directed at those studies that allege to support the claim that these dolls can be useful for assessing sex abuse, such documentation would be very difficult to provide at this point. In the state of California the dolls are not considered to satisfy the Kelly-Frye test for admissibility, which requires that the procedure has been generally accepted as reliable in the scientific community in which it was developed. This decision was based on the weaknesses of the studies supporting their use (Gardner, 1991a; Underwager and Wakefield, 1990; Wakefield and Underwager, 1988).

The American Academy of Child and Adolescent Psychiatry (AACAP) (1988) has also issued a position statement on the use of these dolls. The AACAP's position is similar to that of the APA's, but is a little stronger with regard to the caveats about the dolls' use. The AACAP makes direct reference to the fact that findings based on the dolls are not admissible as evidence in the state of California.

Even those who are the strongest supporters of the use of the dolls generally agree that ``leading questions'' should be avoided as much as possible because they can produce responses in the child that may suggest sex abuse when it has not occurred. However, those who warn against the use of leading questions often use them to a significant degree in their protocols. For example, White et al. (1985) warn, ``The interviewer should pose questions in a nonleading fashion.'' Yet, in the same protocol they list 14 questions, each of which most would consider to be highly leading, e.g., ``Have you been touched on any part of your body?'' ``Has anyone put anything on or in any part of your body?'' ``Has anyone else asked you to take off your clothes?''

Boat and Everson (1988) exhibit the same disparity in their protocol regarding the caveat against using leading questions. The authors state, ``Background information should therefore be used only for guiding the conversation (e.g., `Your Mommy told me you visited Uncle John last week.') and not [except in rare circumstances to be outlined later] for asking the child questions that may be leading or suggestive (e.g., `Is Uncle John the one who hurt you?').' The authors do not appear to appreciate the fact that the statement: ``Your Mommy told me you visited Uncle John last week'' is very leading. Perhaps it is a little less leading than ``Is Uncle John the one who hurt you?'' but it is nevertheless leading in that it directs the child's attention to the visit with Uncle John the previous week, the visit during which some sexual encounter is alleged to have taken place. Furthermore, in spite of this caveat, the protocol itself includes a series of questions, each of which is highly leading [in each of the examples to be given the line ( ) represents the name the child uses for that particular body part], for example, ``Has anyone touched your ? ``Have you ever seen anybody else's ?'' ``Has anyone asked you to touch their ?'' and ``Has anyone taken pictures of you with your clothes off?'' Most examiners, I believe, would agree that these questions are highly leading.

Obviously, we have a problem here regarding the exact definition of a leading question. Providing a definition for this term is one of the objects of this chapter. However, I will go beyond this issue and discuss the concepts of what I refer to as ``leading stimuli'' and ``leading gestures.'' The introduction of these terms is important if one is to appreciate more fully what is actually occurring when examiners use anatomical dolls. As discussed elsewhere (Gardner, 1987, 1991a), in the early 1980s I began seeing the introduction of false sex-abuse allegations in the context of highly contested child custody disputes. This allegation proved a valuable vengeance and exclusionary maneuver. In the last few years I have also been involved in day-care center cases in which sex abuse has been alleged. In association with my involvement in both of these areas, I have had the opportunity to review approximately 300 hours of videotapes of examiners using anatomical dolls in the course of their interviews. These interviews were not simply conducted in the Greater New York City area (where I live and work), but in other parts of the country as well. Accordingly, I have had the opportunity to observe a broad spectrum of interviewers. These tapes have not been viewed leisurely; rather, they have been carefully scrutinized--most often with a sentence-by-sentence analysis--with particular attention to the examiner's gestures, intonations, and other details that might not be given primary consideration by many observers. One of the unanticipated outcomes of such viewing has been the consistency and the similarities of their approaches, which appear to have been influenced significantly by the procedures promulgated by White et al. (1985), White and Halpin (1986), and Boat and Everson (1988). I have not only noted the frequent use of leading questions, even more extensively than used by the aforementioned authors, but the introduction of what I refer to as leading gestures. Although leading questions can easily be seen on the transcripts of these interviews, the leading gestures are rarely described by the transcriber. These gestures play an important role in what is actually taking place during these interviews and, I believe, they play a role in the ``programming'' process that is occurring with the utilization of these dolls. I will also utilize the term leading stimuli, which refers to dolls (especially anatomical), body charts, and other instruments that can serve to contaminate the interview, especially in the direction of drawing the child into talking about sexual issues and contributing thereby to the brainwashing process.

My main purpose is to discuss in detail these three areas of importance in sex-abuse evaluation interviews: (1) leading stimuli (of which anatomical dolls represent one example), (2) leading questions, and (3) leading gestures. Defining these concepts more accurately will, it is hoped, improve communication among examiners, especially with regard to what is and what is not a leading question. It is important that the reader not view these concepts simply as yes/no phenomena, whether present or absent; rather, each is best understood as lying on a continuum from not being present at all to being utilized to a significant degree. For each of the three concepts I will demarcate four points along this continuum: not present (that is, absent), minimally present, moderately present, and maximally present. However, I wish to emphasize that these are merely guideposts for categorization. The general principle that is followed for ascertaining which level is most appropriate is the degree to which the stimulus shrinks the universe of possible responses by the interviewee. The less the shrinkage, the greater the likelihood the stimulus will fall at the ``absent'' end of the continuum. And the greater the shrinkage or narrowing, the greater the likelihood the stimulus would be considered to be ``maximal'' in this regard.

Last, I will not make a distinction between leading questions and ``suggestive questions.'' Although some may make a distinction between the two, I will not do so, nor will I use the term. Rather, I will use the term leading questions to refer to all types of questions that have the capacity to suggest an answer and utilize terminology that attempts to refine the various degrees of suggestibility within the question. Similarly, I will not make any differentiation between nonleading questions and ``open-ended questions.'' I consider open-ended any question that does not contain within it a contaminating focus for response. My comments here will refer to the various degrees to which a question is open and utilize the terminology herein, which more specifically attempts to define the degree of ``open-endedness.''


The stimuli that I will be focusing on here are instruments that are generally used in interviewing by mental health professionals, especially psychologists and psychiatrists. They are instruments that have been designed to elicit verbal responses by the interviewee, especially responses that reveal psychodynamic information and personality manifestations. They all involve an external visual stimulus that serves as a focus for the interviewee's verbal response. They differ from leading questions in that the stimuli in the leading question category are entirely verbal. They differ from leading gestures in that the leading stimuli category includes standard, well-defined materials that are easily recognized as providing visual stimuli. Leading gestures, in contrast, although visual, are an incidental form of visual stimulation that may not be directly attended to by the interviewer or the interviewee but have a profound effect on the process of the interview.

Nonleading Stimuli

The ideal nonleading visual stimulus would be no stimulus at all. The psychoanalyst's blank screen for free associations would be an example. The blank card of The Thematic Apperception Test (TAT) (Murray, 1936) would also be a good example. Whereas all the other cards of the TAT depict a particular scene with a specific number of individuals, the blank card is completely blank. The universe of possible stories that the interviewee can provide is basically infinite and the external facilitating stimulus does not provide anything that could reasonably contaminate the associations and thereby ``shrink'' the universe of possible responses.

Minimally Leading Stimuli

A good example of minimally leading stimuli are the Rorschach cards (Rorschach, 1921). These inkblots were selected because they have only limited similarity with known objects. There are some cards, however, that vaguely resemble known objects such as bats, butterflies, and ``two drummers drumming.'' Surprisingly, these are referred to as the ``popular responses'' and are not particularly indicative of psychopathology. In fact, not to see these popular responses may suggest the presence of such pathology. Because most of the stimuli are not suggestive, and because the universe of possible responses has not been significantly reduced by these stimuli, I place these cards in the the category of minimally leading stimuli. Another instrument in this category is The Storytelling Card Game (STCG) (Gardner, 1988a). In this psychotherapeutic game (which can also be used as a diagnostic instrument for eliciting psychodynamic material), there are 24 scene cards, 4 of which are blank. The nonblank scenes depict scenes only and do not contain any figures, human or animal. Some typical scenes: forest, child's bedroom, city street, farm, etc. The child is asked to select one or more figurines (from an array of 15, ranging in age from infancy to old age), place them on a card, and tell a self-created story. It is almost as if the child is creating his (her) own TAT picture card. The blank card pictures used in the STCG are somewhat more contaminating than the blank card of the TAT because in the former there is an actual visual stimulus (the figurines placed on the blank card), whereas in the latter there is none. Because the children create their own pictures, the STCG pictures are less contaminating than the standard nonblank TAT cards (to be described in the next category).

The instruments used in this category have limited capacity to constrain or contaminate the universe of possible responses. They are well viewed as catalysts for projections created by the interviewee, with little potential by the external facilitating stimulus to modify significantly the interviewee's inner psychological processes.

Moderately Leading Stimuli

In this category the external visual stimuli are quite specific; yet they allow for a universe of possible responses. However, because of the specificity of the external visual eliciting stimuli, the universe of possible responses is smaller than that provided by the stimuli in the previous two categories. Most of the TAT cards fall well into this category. In each of the cards one can identify easily the number of individuals depicted, the sex, and even something about the attitudes or emotions of the individuals depicted (although these tend to be somewhat vague). Some examples: Card No. 1 depicts a boy looking down at his violin, which is lying on the table in front of him. Card No. 2 depicts a school girl (carrying books) behind whom is a farm scene in which an adult female, an adult male, and a horse are clearly indicated. Card No. 5 depicts an adult woman standing at a doorway looking into a room (which could be either a foyer or a living room). Card No. 17BM depicts a naked man climbing up a rope. Obviously, the TAT cards shrink the universe of possible responses much more than the stimuli in the aforementioned two categories. Yet, there is still a universe of possible responses to each of these cards.

Maximally Leading Stimuli

The materials described in this category shrink significantly the universe of possible responses. They provide a highly specific stimulus that is very likely to draw the interviewee's attention directly to it. Generally, they provide stimuli that are likely to bring about strong emotional responses in the interviewee. Many of the cards in the Children's Apperception Test (CAT) (Bellak and Bellak, 1949) are in this category. Some examples: Card No. 1 depicts three chickens sitting around a table, in the center of which is a bowl of food. In the background is the adult chicken (most likely the mother). Card No. 7 depicts a tiger leaping menacingly toward a small monkey. His mouth is open. He is approximately ten times the size of the monkey. Card No. 10 depicts a bathroom scene in which there is a toilet. Sitting nearby is a mother dog spanking a baby dog.

The Children's Apperception Test-Supplement (CAT-S) (Bellak and Bellak, 1963) also contains cards that warrant placement in this category. Some examples: Card No. 5 depicts a kangaroo walking with crutches. Both his (her) left leg and tail are bandaged. Card No. 10 depicts a female cat, obviously pregnant. Because she is standing up, her abdomen is easily recognized as protuberant. TAT Card 8BM also warrants placement in this category. It depicts a man being operated on by two other men, one of whom is holding a knife (or scalpel). Next to him is a gun. In front of all of this is a young man looking out at the viewer. The presence of the gun, knife, and surgical operation warrant this card's being placed in the maximally leading stimulus category because it provides a much greater shrinkage of the universe of possible responses than is provided by the aforementioned TAT cards, which I considered to warrant placement in the moderately leading stimuli category.

The centerfold of many pin-up magazines warrant placement in this category. Whatever the nature of one's reactions (and I admit to a wide variety), there is no question that such a picture provides very compelling stimuli. Groth (l984) has introduced anatomical drawings that are often used in the assessment of children who are being evaluated for sexual abuse. Their purported purpose is to enable the examiner to learn exactly what terms the child uses for the various body parts. The child is presented with cards on which are depicted naked people. Both anterior and posterior views are presented. I have serious reservations about these cards. They provide maximal leading stimuli and communicate to the child that sexual issues are proper and acceptable topics to focus on. If this were indeed the only example of such stimuli used by the examiner, the cards probably wouldn't be too high a contaminant. However, when used along with anatomical dolls, leading questions, and an atmosphere in which the examiner communicates to the child that sex talk is the ``name of the game in this office,'' they may contribute to the creation and/or promulgation of a false sex-abuse accusation. Their ostensible use for learning the names the child uses can easily be viewed as the examiner's rationalization for sexualization of the interview. The examiner could achieve the same goal by simply asking the mother, outside of the child's presence, what names the child uses for the various body parts. In this way, the examiner would gain the benefits to be derived from having this information without contaminating the interview in the sexual direction.

I consider anatomical dolls to be in this category of maximally leading stimuli. They provide highly compelling and unusual stimuli that cannot but draw the child's attention from other issues that may have been on his (her) mind. Some of these dolls have organs that are disproportionately large; others do not. But in both cases the dolls are unusual and thereby attract extra attention. They are not the kinds of dolls that the vast majority of parents (including evaluators who are parents) are likely to give their children as birthday or holiday presents. If a rambunctious teenager were to surreptitiously pencil in genitals, breasts, and pubic hair onto a set of TAT cards, the examiner would probably consider them ``ruined'' and be quite upset at the miscreant. Such alteration of the cards obviously makes them even more contaminating and would put them in the category of a pin-up magazine centerfold.

I wish to emphasize that my experiences with viewing the videotapes of examiners who utilize these dolls is that leading questions and leading gestures are greater contaminants than the dolls, their strong contaminating potential notwithstanding, and this is one of the reasons why many of the studies that conclude that the dolls are not contaminating are not reproducing the actual situation in which they are commonly used. Specifically, these studies are usually done under laboratory conditions in which the examiners attempt to reduce significantly the use of leading questions because of the appreciation that they have the effect of biasing a sex-abuse evaluation. A typical study involves a child being allowed to play along with the dolls. If an examiner is present, he (she) says nothing or makes a few catalytic statements. Others generally observe the child through a one- way mirror. Activities or comments that could be construed as sexual are quantified.

The main problem with these studies is that they do not accurately reproduce the conditions out there in the real world where examiners are literally bombarding children with leading questions in association with their use of these dolls. Those that appear to support the conclusion that the dolls are not contaminating usually include a few nonabused children who, indeed, provided responses similar to those one obtains from children who have been genuinely abused. This, of course, argues against their ``safety.'' Of relevance here is the common finding that the majority (usually 80-95 percent) of children do not provide responses suggesting sex abuse. This figure is used to justify the conclusion that the dolls are safe. However, the same studies indicate 5-20 percent false positives, i.e., the dolls facilitating responses that are the same as those seen in children who were genuinely abused. From the point of view of an accused person, these false positives indicate that the dolls are very risky and even dangerous and could contribute to the incarceration of a person who did not sexually abuse the child being evaluated. The situation is similar, I believe, to the one in which a pharmaceutical company would propose placing on the market a drug that they claim is ``only 5-20 percent lethal.''

The reader who is interested in reading directly the studies that come to the aforementioned conclusions does well to refer to the articles by Gabriel (1985), White et al. (1985), Jampole and Weber (1987), Sivan et al. (1988), Glaser and Collins (1989), Clarke-Stewart et al. (1989), Thompson et al. (1991), August and Forman (1989), McIver et al. (1989), and Realmuto et al. (1990). Moreover, Yates and Terr (1988a, 1988b) debate the anatomical doll issue and present the common arguments provided by both those who approve of and those who object to the dolls. At this point, I believe that the aforementioned studies do not provide convincing evidence that anatomical dolls are particularly safe. They indicate that the dolls--even when used alone, without leading questions and leading gestures--play a role in bringing about and/or promulgating false sex-abuse accusations.


I use the term leading questions to refer to questions that engender in the mind of the listener a specific visual image that is not likely to have been produced had the question not been asked. It usually induces the formation of a visual image that may serve as a reference point for future responses and behavior. When a leading question is asked, the examiner cannot be sure whether the interviewee's response has been suggested by the leading question or whether it would have been provided anyway. Such ``seed planting'' interferes significantly with the data- collection process, the purpose of which is to ascertain whether something really happened (such as is the case in a sex-abuse evaluation). Leading questions can also contribute to the brainwashing process that can take place in the course of an evaluation. Engendering as they do fantasies of events that might not have occurred, there is the risk that the imagery that leading questions induce will come to be believed by the interviewee. Even adults often have problems remembering the original source of information or a recollection. A visual image can come to have a life of its own, especially if repeatedly suggested and brought into conscious awareness. And children, being more suggestible, are more likely to forget the source of an engendered visual image and are more likely to come to believe that the events portrayed therein actually occurred.

Consider, for example, a three-year-old girl who has never performed fellatio on her father, was never asked to do so by him, and never even entertained the notion of doing so. An examiner asks this child, ``Did your father ever put his penis in your mouth?'' The question engenders in the child's mind the visual image of her performing fellatio on her father. It is probable that this particular thought never previously entered her mind. The question, however, creates an image of just such an event. Her initial response may be one of denial and possibly even some revulsion. A few days later, when asked the same question (by the same or another examiner), the fellatio-with-father image is then brought into conscious awareness. However, this time she may not be able to differentiate between a recollection derived from the question asked a few days earlier and images related to an actual event. At that point she may respond with, ``I think so-- I'm not sure.'' If the examiner zealously wants the child to respond in the affirmative (a not uncommon situation, in my experience), he (she) may pursue the issue with statements such as, ``Think hard,'' ``You can tell me,'' ``It's safe here,'' and ``I'll make sure he won't be able to do it again.'' If one adds here the child's desire to ingratiate herself to the examiner, who she recognizes wants the answer to be yes, it is more likely that the child will consider the ``recall'' of the visual image as a manifestation of her having indeed had such an experience. It is for these reasons that leading questions can result in a sex-abuse evaluation that is much more a learning process than an exploratory one. Leading questions, then, play a role in what is often criticized as a ``brainwashing'' or ``programming'' process.

Leading questions, like leading stimuli and leading gestures, are on a hierarchy from questions that are not leading at all to those that are maximally leading. There are differentiating criteria that delineate the subcategories, but the reader does well to consider there to be a continuum, with the subcategories serving as guidelines for categorization. First, I focus primarily on the verbal communications of the evaluator (auditory stimuli), and I do not direct myself to visual environmental stimuli (including leading stimuli and leading gestures) that are usually operating simultaneously and also have an effect on the child's productions.

Nonleading Questions

Here, again, the psychoanalytic model can be useful. Although there have been many serious criticisms of Freudian psychoanalysis (and I agree with many of them), I do believe that the blank-screen principle is a valid one. It is the best setting for enabling the therapist to find out what is on the patient's mind, uncontaminated by comments made by the therapist. In the ideal psychoanalytic situation, the session begins without the therapist saying a word. Whether the therapist is behind the couch or face to face is not relevant here. What is relevant is that there is practically no direct verbal communication originating with the therapist. The patient recognizes that he (she) will be expected to start talking about the things that are coming into conscious awareness. Thus, a completely nonleading question, in its purest form, is no question at all. There are opening questions, however, that justify placement in this category. Questions like ``So what's on your mind?'' and ``What would you like to talk about today?'' still provide for a universe of possible responses. Potential contaminants provided by these questions are highly unlikely.

Young children, especially under the age of five or six (the age range in which anatomical dolls are most frequently used), do not generally respond well to such questions as ``What's on your mind?'' and ``What would you like to talk about today?'' However, after an initial ``Hello'' and ``Please come in'' (comments that, in my experience, are genuinely nonleading), the child might be observed by the examiner to see what he (she) will do. The examiner can sit and say absolutely nothing and will sometimes be successful in getting the child to spontaneously verbalize, especially after selecting some object in the room to serve as a focus. However, once this object (often a toy) is picked up, we no longer have a blank screen. (We have, now, a category of leading stimulus.) If, however, the child starts talking spontaneously, then we do indeed have a nonleading- question-type situation. Sometimes the child's talking can be facilitated by the child's parent(s) coming in initially, especially at the beginning of the first interview. The parent's presence, obviously, will make the child more comfortable in a strange situation and increases the likelihood that the child will verbalize. However, in order to ensure that the child's verbal productions will not be contaminated by their comments and/or gestures (no matter how subtle), they must be strictly instructed to say nothing at all. If one is successful in this regard (often the case, in my experience), then one is likely to have a situation in which the child begins to talk spontaneously. Once the child is comfortable, the parent(s) can be asked to leave the room, and then one is more likely to have a setting in which spontaneous verbalizations will come forth. Last, the parents need not be brought into the second interview because of the child's previous familiarization with the examiner and his (her) procedures.

Minimally Leading Questions

These are questions that narrow the universe of possible responses to a very limited degree. They are questions or openings that are rarely used as foci for the interviewee's response. An example would be a situation in which a evaluator provides the child with a wide assortment of materials (dolls, farm animals, zoo animals, drawing paper, crayons, clay, doll house) and simply says to the child, ``You can play with anything you would like to play with here.'' Then, while the child is playing, the evaluator sits silently, awaiting the child's spontaneous comments and verbalized fantasies. There has been some minimal contamination here because the universe has been narrowed somewhat by virtue of the fact that the child is being asked to direct his (her) attention to the play material. However, that universe has not been narrowed significantly because there is still a large universe of possible comments the child might make while so engaged. If the child doesn't say anything, the examiner might say, ``What are you thinking about while you're playing with those things?'' Again, there is a large universe of possible thoughts the child might have. Here, the examiner might say, ``I'd like you to tell me a story about the things you're playing with.'' Here, too, the universe is not significantly constricted. Children under the age of five, however, are generally not cognitively capable of providing well-organized stories--especially with a beginning, middle, and an end. However, they often will, in response to this question, provide a string of loosely connected associations. (They are likely to comply because of the aforementioned desire to ingratiate themselves to adult authority.)

The same principle is applicable in the introduction to the mutual storytelling technique (Gardner, 1968, 1971, 1992a). The introductory material provided by the therapist in the ``Make-Up- a-Story Television Program'' is general, nonspecific, and does not include material that has any particular ``pull'' toward a particular story theme. The examiner's responding story (which is therapeutically designed to include the same characters in a similar setting, but incorporates healthier modes of adaptation than those utilized by the child) does provide highly specific material. However, the response here is a therapeutic message, derived from the child's uncontaminated fantasies, and is not presented as a noncontaminated projection. (The therapeutic utilization of this game is not the purpose of this chapter. Rather, only the use of the mutual storytelling technique's story-eliciting capacity is being focused on here.)

Many (but not all) of the cards in The Talking, Feeling, and Doing Game (Gardner, 1973) warrant placement in this category. This board game, which was designed to elicit psychotherapeutically valuable information from children in the context of a mildly competitive board game, involves each player rolling the dice and moving a playing pawn around a curved path from start to finish. When the player's pawn lands on a white square, a Talking Card is taken, on a yellow square, a Feeling Card is taken, and on a pink square, a Doing Card is taken. When a player answers the question or responds to the instructions on the card, a reward chip is obtained. Obviously, the object of the game is not simply to earn chips and acquire a prize, but to use the cards as points of departure for psychotherapeutic interchanges between the therapist and patient. Examples of cards that warrant placement in the minimally leading questions category: ``Make up a dream,'' ``Someone passes you a note. What does it say?'' ``Make believe you're looking into a crystal ball that can show anything that's happening anywhere in the whole world. What do you see?'' As can be seen, there is indeed a universe of possible responses to each of these questions. Although not as large as the blank screen universe provided by the facilitators in the category of completely nonleading questions, they are certainly close to this ideal.

Moderately Leading Questions

Questions in this category narrow somewhat the universe from which the questions in the previous category draw; however, the constriction of the universe is not that great that significant idiosyncratic material cannot be obtained. Most of the cards in The Talking, Feeling, and Doing Game (Gardner, 1973) warrant placement in this category, for example, ``If the walls of your house could talk, what would they say about your family?'' ``Everybody in the class was laughing at a girl. What had happened?'' ``A boy has something on his mind that he's afraid to tell his father. What is it that he's scared to talk about?'' ``Make up a lie.'' ``What's the worse thing that ever happened to you in your whole life?'' ``All the girls in the class were invited to a birthday party except one. How did she feel? Why wasn't she invited?'' and ``Suppose two people were talking about you, and they didn't know you were listening. What do you think you would hear them saying?'' These questions narrow somewhat the range of responses, but still allow for a universe of possible answers. There is a universe of possible lies that the child could possibly tell. There is a universe of possible reasons why everybody in a class would laugh at a girl. There is a universe of possible things a person could hear others saying about oneself when eavesdropping. Yet, the questions do focus somewhat on a particular area of inquiry, e.g., friendships, lying, antisocial behavior, etc.

When doing an evaluation of children I may ask the following question: ``You know, nobody is perfect. Everybody is a mixture of things you like and things you don't like. Your parents are no exception. I'd like you to tell me the things about your mother and father that you like and the things about your mother and your father that you don't like. Which one do you want to start with?'' There is a universe of possible responses to this question, even though it narrows down to the assets and liabilities of the child's parents. Accordingly, they warrant placement in the category of moderately leading questions.

A verbal projective instrument designed by Kritzberg (1966) also warrants inclusion in this category. The child is asked the question, ``If you had to be turned into an animal, and could choose any animal in the whole world, what animal would you choose?'' After the child responds, he (she) is asked the reason for that choice. Following this, the child is asked for his (her) second and third choices and the reasons why. Then the child is asked what three animals he (she) would not want to be, and the reasons why. A similar series of questions could be asked about objects into which the child could be transformed and, for older children, the specific persons they would choose and not choose to be changed into if they were free to select from the whole array of humanity, living and dead, famous and not famous, real and fictional. Although there is a finite number of animals, objects, and persons (for older children) from which to choose, the number is still quite large (and the older the child, the larger the number). Although the word universe is not applicable here, we still do not have the kind of constriction and seed-planting that one sees in the maximally-leading-question category. Furthermore, once the animal, object, or person has been selected, the reasons for that selection allow for a much larger universe of possible responses. The second part of the question (the ``why'' part) provides, therefore, the more valuable information because it allows for the tapping of a larger universe of possibilities.

It is in this category that we begin to see some of the contaminating leading questions utilized by many sex-abuse evaluators. For example, the examiner might ask, ``Tell me about the school you used to go to.'' On the surface, this seems like an innocuous enough question and would be the kind that many evaluators might ask in the course of their evaluations. However, in this particular case the examiner knows that the school the child used to attend was closed down because the directors allegedly sexually molested the children there. Taking this into consideration, and taking into consideration the fact that this was the first substantive question asked in the interview, we see how the word universe has no applicability here. A response that includes nonsexual events is usually of little concern to such an evaluator. And this is an important point. To the degree that the evaluator is thinking about a particular answer, to that degree the question is likely to be leading. Of course, the inclusion of this criterion makes it more difficult to assess the question on its face value regarding the degree to which it is leading, but it is not usually difficult to speculate about what's going on in the evaluator's mind when every single question is directed toward the alleged sex-abuse event, no matter how unrelated to sex they may initially appear.

I cannot emphasize this point strongly enough. The competent examiner is truly going to be open to any response the child provides and use that as a point of departure for further inquiry and elaboration, whether it be for evaluative or therapeutic purposes. This is the opposite of the approach utilized by most of the sex-abuse examiners whose reports I have read and/or whose videotapes I have studied and analyzed. It is clear that they start off from the position that the sex abuse is most likely to have taken place and their job is merely to fill in the details. Noncorroborative data are ignored, denials are rationalized away, and one knows then that ostensibly minimally leading questions such as ``Tell me about the school you used to go to'' are actually quite contaminating and warrant placement in a higher category of leading questions.

Another example: ``Tell me about your Uncle Bill.'' Again, one must consider this statement at two levels. Ostensibly, it may not be highly leading, although it is intrinsically leading under the best of circumstances because the focus is on one person of the 5.3 billion people on earth. However, there is still a universe of possible statements one could make about Uncle Bill. In this case, however, Uncle Bill is the alleged perpetrator. We know, also, that he has been selected from the wide assortment of the child's friends and relatives who the examiner could have chosen to focus on. We know, then, that the examiner is primarily interested in discussing the alleged sexual encounters between the child and Uncle Bill, even though the examiner may be professing a completely neutral position regarding whether sex abuse took place--the usual position of many evaluators (especially ``validators''), professions of neutrality notwithstanding. Accordingly, it has a ``seed-planting'' effect in that the child gets the message that the examiner is interested in talking about issues related to sex abuse, a subject selected from the universe of issues that could be focused upon in the examiner's office.

Maximally Leading Questions

A maximally leading question is one about which there is no doubt what the examiner is interested in discussing and/or hearing. It is one star pulled out from a million galaxies. Many are totally innocuous, e.g., ``How old are you?'' ``What grade are you in?'' and ``What's your teacher's name?'' These are highly selected questions, point out exactly what the examiner is looking for, and lead the child down a very specific path. Such questions may even have the fringe benefit of relaxing a child during his (her) first interview with a strange therapist. The question ``What's the name of your school?'' might very well be in the same category. However, if the school is the place where the alleged sex abuse took place, and if the school question is one of the first the examiner asks, then the same question now falls into the moderately leading question category (for reasons described above). If, however, the examiner asks, ``Tell me exactly what your teacher, Mr. Jones, did to you after he pulled down your pants,'' we have a statement that warrants placement in the maximally-leading- question category. It evokes a specific image of Mr. Jones pulling down the child's pants. Mr. Jones may or may not have actually pulled down this child's pants. Once the question is asked, however, the likelihood of our knowing whether or not this actually happened (at least with the child as a source of information) is reduced significantly. (And this problem is compounded by examiners who see only the child and then come to a conclusion regarding whether Mr. Jones perpetrated an act of sexual abuse.)

Many of the questions in the White et al. (1985) and Boat and Everson (1988) protocols warrant placement in this category. For example, the question ``Did anyone ever touch you in the wrong place?'' engenders the specific fantasy of the child's being touched sexually, whether or not the child ever was. The question ``Has anyone ever taken your picture?'' seems innocuous enough and, one could argue, certainly does not warrant placement in the maximally-leading-question category. However, we know that the examiner who asks this question is not thinking of pictures taken at the zoo, on picnics, or at amusement centers. Rather, the examiner is thinking about child pornography, and the visual image in the examiner's mind is of an adult with a camera taking pictures of one or more naked children, possibly engaged in a wide variety of sexual activities. This is confirmed by the fact that the question is one of a series, all of which relate to sex abuse. If the child were to respond, ``Oh, yes. My daddy took a lot of pictures of me when we visited Disney World,'' that would not be considered a satisfactory or acceptable answer. In all probability the examiner would then ask about pictures taken in other settings in the hope that a sex-abuse scenario would be described. Or, the examiner might get more specific and ask leading questions about whether the child was naked in the pictures, questions about a particular person (especially the alleged perpetrator) taking pictures, etc. Another example: ``Have you ever been without your clothes?'' If the child were to say, ``Of course, when my mommy gives me a bath, I don't have clothes on,'' the examiner is not likely to find this an acceptable answer. He (she) is likely to ask more questions about other people, especially the alleged perpetrator.

Many yes/no questions fall into this category. For example, ``Did anyone ever touch your private places?'' serves as an entree into a specific discussion about sex abuse. Whether the child says yes or no does not usually provide useful information. If the child says yes, one doesn't know whether the child is saying so in order to ingratiate himself (herself) to the evaluator or because the child really had such an experience. If the child says no, one doesn't know if the no answer relates to reality or if the child had such an experience and is too ashamed, guilt-ridden, fearful, etc., to answer in the affirmative. But less important than the significant drawbacks of the yes/no question (and lawyers have yet to learn this) is the fact that the question ``plants a seed'' because it induces a fantasy of a child's being touched in ``private places,'' whether or not he (she) has had such an experience. And this is one of the ways that legal interrogations (with their frequent use of yes/no questions) program children into believing they have had such experiences, when they did not.


This is a most important area that has not been given proper attention by those involved in sex-abuse evaluations. Written transcripts of videotapes rarely mention the gestures of the evaluator. Only the verbalizations are usually recorded, unless something very unusual and dramatic has taken place. Audiotapes, obviously, tell us practically nothing about gestures and other body movements. Yet, throughout much of the interview, the child is looking at the evaluator and obviously is being affected by what he (she) sees. Children model themselves after adults, especially their parents and other authorities. The emulation process often involves the notion: ``If I act like him (her), he (she) will like me better.'' The imitation factor contributes to children's learning about how to function in the adult world. There is a reflexive need on the part of children to imitate significant adults. It relates to the need to learn the acceptable patterns of behavior that will enable them to be like others and thereby fit in well with and be accepted by the group. I call it the ``keeping-up-with-the-Joneses'' phenomenon.

Another factor operative here is that of sanction. If the child observes the evaluator to be performing an act that might generally be considered unacceptable (for example, placing a finger in a doll's vagina or stroking a doll's penis), he (she) is more likely to repeat that act, even if aware of its unacceptability. The child often operates on the principle: ``If it's okay for him (her) to do it, it's okay for me to do it also.''

Again, I divide leading gestures into four categories. Again, we are dealing here with a continuum: from nonleading gestures at the one end to maximally leading gestures at the other, and the categorization is primarily for the purpose of providing guideposts.

Nonleading Gestures

Once again, we start with our old friends, the classical psychoanalysts, many of whom (but certainly not all) are experts at doing nothing. (The reader should appreciate that I myself am a trained analyst and so my criticisms of these people are not based simply on books I have read on the subject.) One of the reasons given for sitting behind the couch is to lessen the likelihood that the patient will be affected by the psychoanalyst's facial expressions and gestures. Although I think much is lost by conducting therapy from this position, especially the human relationship (Gardner, 1988b, 1992b), these people do have a point. Although the psychoanalyst sitting behind the couch provides one of the best examples of nonleading gestures, most children who are being evaluated for sex abuse are not coming to be analyzed and most children who are in analysis do not have analysts who work from behind the couch. Nor am I recommending that sex-abuse evaluators buy couches. The evaluator, however, can indeed avoid leading gestures, gestures that provide communications that will contaminate the blank- screen interview. To ask a nonleading question without any contaminating facial expressions or gestures is easily accomplished. One can say to a child, ``What would you like to do here today?'' without any kind of directive movement or glances. Also, after asking nonleading introductory questions in the name-rank-and- serial-number category, the examiner just might sit, somewhat expressionless, and watch the child to see where he (she) will gravitate and wait until the child begins to speak.

Minimally Leading Gestures

These gestures, like most of the gestures I discuss here, are usually associated with verbalizations that may or may not be in the same category regarding its contamination of the blank screen. If the examiner wishes to direct the child's attention to the toys, because they are likely to provide catalyzation of the expression of the child's naturally occurring fantasies, he (she) might, with an arc sweep of the hand across the toy shelves, say, ``You can play with any of the toys in the room.'' This gesture is selective in that it strongly suggests that the child confine himself (herself) to the toys and not to other objects in the room. It restricts somewhat the universe of possible activities and statements, but it is minimally contaminating. The child still has the option to select any of the toys and if they are intrinsically of low-contamination potential, then the gesture will have served its purpose.

The examiner might take a pad of drawing paper and crayons, put them down in front of the child, and say, ``I'd like you to make a drawing. Any drawing at all. Draw anything in the whole wide world that comes to mind. Then, when you finish, I'd like you to tell me about what you've drawn.'' This can be said while pointing to the crayons and blank sheet of paper. These gestures reinforce the verbal request and increase the likelihood that the child will provide a reasonably pure projection. The same principles hold when one is instructing a child to draw a figure for the Draw-a-Person Test, the Draw-a-Family Test, or the request that the child play with dolls, clay, fingerpaints, sand, or other traditional play therapy materials. (My assumption here is that the examiner recognizes the contaminating effects of such games as chess, Monopoly, Candy Land, etc. Unfortunately, there are many therapists who do not appreciate this obvious fact.)

Winnicott (1968) utilizes a game that he refers to as ``squiggles.'' The therapist or the child begins by drawing with pencil on blank paper a nonrecognizable scribble (``squiggle''). The other party then pencils in additional lines and curves. Back and forth they go until the child reaches the point where he (she) recognizes some identifiable figure. Then the figure serves as a point of departure for either storytelling or other therapeutic communications. The gestures here are basically of limited contaminating potential. They contribute to the drawing of a figure of varying similarities to actual objects. However, the child is a contributor and the identification of the figure so drawn is based on what the child sees in it, i.e., the child's projections, rather than what the examiner considers the squiggle to look like. Because there is still a large universe of possible associations and stories to the squiggle, the game warrants categorization in the minimally- leading-gesture category.

Moderately Leading Gestures

Evaluators who use anatomical dolls will commonly say to children, ``Show me how he did it with the dolls.'' The ostensible purpose here is to facilitate the child's providing an accurate description in order to compensate for the child's immature level of verbal and cognitive communicative capacity. Another argument given for their utilization is that they help inhibited children talk about sex abuse. Although there may be some minimal justification for these arguments for the dolls being used, the encouragement of the physical enactment tends to entrench the child's belief in the events so portrayed, whether or not they actually occurred. When this direction is given, the evaluator often picks up the dolls and hands them to the child, again strengthening the power of the verbal request.

An example of a moderately leading gesture would be the examiner who says, while picking up a family of puppets, ``Let's play with these hand puppets. I'll take the mother and father (places them on his [her] hands) and you take the boy and the girl and put them on your hands. Now what happens?'' Under these circumstances the child is likely to start moving his (her) hands and verbalize some kind of activity while engaging the therapist's puppets in a similar activity. If it was the therapist who suggested the game, then the universe of possible associations has been significantly contaminated. But even if the child suggests the game, the game in itself has limitations imposed upon it by the physical structure of the puppets and the hands of the humans who are playing with them. Children commonly will bang the heads of the puppets, one against the other. The examiner who interprets this to mean that the child is angry may be stretching a point significantly. The dolls almost ask to be banged together or to engage one another in various kinds of hostile play. Accordingly, the therapist's leading gestures here bring the child down a somewhat narrow path, the path of hostile play.

Levy (1939, 1940) describes a type of child psychotherapy that he refers to as ``release therapy.'' This method was designed to help children verbalize their thoughts and release their feelings about traumatic situations. In order to facilitate this process he structured the doll play in such a way that the child is likely to talk about a particular situation, especially a traumatic one. For example, if a child was dealing with the trauma of being in a hospital, he might walk a boy doll into a make-believe hospital room, lay the child down in a hospital bed, and ask the child to talk about what's happening to the little boy. The child who is reacting to the birth of a new sibling might be confronted with a structured doll-play situation in which the mother is breastfeeding the baby doll. The little boy doll is walked into the room, looks at the mother and baby, and the patient is asked to describe what the little boy is thinking and feeling. This game clearly involves the therapist in moderately leading gestures. Its risk is that it will pressure children into talking about issues that they may wish to repress or that they may be too anxious to discuss. It also has the drawback of ``muckraking'' and bringing up past traumas that have already been put to rest in a way that is not pathological.

The principle here is that the more intrusive the therapist is, the greater the likelihood of brainwashing and programming, and the greater the likelihood the therapy will produce psychological trauma. And this is what happens, I believe, with children who have never been abused and who have been ``treated'' for an abuse that never took place. They are made to believe that they were abused. The effects of such programming have yet to be studied because these children represent what is now a new form of psychopathology (Gardner, 1991a).

Many evaluators are advised to ascertain, early in the first interview, the terms the child uses for the various body parts, especially the sexual ones. As mentioned, the ostensible purpose here is to enable the examiner to communicate better with the child and to understand exactly what organs are being referred to when the child mentions these in the course of the evaluation. This is traditionally accomplished by the examiners pointing to sexual parts on pictures of naked people or special dolls. Commonly, anatomical dolls are used for this purpose, although body charts [such as those of Groth (1984)] are used. Typically, the examiner first points to the various body parts on the clothed doll, e.g., while pointing to the nose the examiner asks, ``What's this called?'' After a few body parts are identified, the clothes come off. Usually the examiner undresses the doll or will ask the child to assist. Once naked, the examiner routinely proceeds to point to the various body parts (breasts, bellybutton, penis, vulva, buttocks, anus) and asks the child specifically to name the part pointed to.

The message being given here is that it is perfectly acceptable, proper, and even desirable to point to and even touch these organs. This is a very unique situation for the child. If a teacher were to do this, the Board of Education might unanimously vote to discharge him (her). If parents, relatives, or neighbors were to do this, they would be suspect as child abusers. It is important to appreciate that there is not only a verbal question being asked here when the examiner touches the doll's penis and asks, ``What's this called?'' A physical activity is being performed which basically sanctions such pointing and touching for the child. I have seen videotapes of examiners who, when asking this question, will take the penis between their thumb and forefinger and talk while manipulating it. This, too, communicates to the child that this kind of activity is acceptable to the evaluator and that if the child would like to act similarly with the doll's penis, he (she) is free to do so. This same principle operates even more poignantly when the examiner puts his (her) finger in the doll's vagina or anus and asks the question, ``What's this called?'' These same examiners will then consider the child's physical activities with the anatomical dolls to be an exact reenactment of their own sexual encounters.

Maximally Leading Gestures

In this, the most contaminating category, the examiner uses the dolls to enact a sexual encounter that has not been enacted by the child and may not even have been described by the child. An example would be of the examiner who takes his (her) finger, inserts it into the doll's vagina, and says, ``Did your grandpa put his finger in here?'' The child, then, by the processes enumerated above, is likely to engage in this very same behavior with the doll, either in that session or in a subsequent session (either with the same examiner or another evaluator). This is an activity that may never have entered the child's mind previously. Talking about it provides one level of implantation of a visual image of this particular experience. Combining the verbal question with an actual physical enactment provides specific visual details that become incorporated into the mental image of the activity. This serves to enhance the likelihood of a visual image forming, an image that will be referred to in future sessions. And, if the examiner believes that what the child does with the dolls is a true reenactment of actual experiences, then a false sex-abuse accusation is likely to be supported or even created.

In one videotaped evaluation I observed, the examiner was convinced that the alleged perpetrator had taken photographs of the alleged victims of sex abuse in a nursery school. The examiner placed the naked anatomical doll on its back, pulled up the legs, and then asked the child, `Did he take a picture of you while your legs were up like this?'' When the child did not respond affirmatively, he placed the doll in the knee-chest position and asked the question again. This was repeated with other positions, many bizarre, to all of which the child responded positively. In one case I was involved in, the examiner pointed to the vagina of a picture of a naked woman and asked the child, ``Did he touch you right here?'' The transcript did not in any way indicate that the examiner was pointing to the vaginal area of the picture. Not surprisingly, the child not only answered in the affirmative, but the examiner used this response as evidence that the child had been sexually molested.

On many occasions I have seen transcripts that simply indicate that the examiner is saying, ``Show me on the girl doll where he put his penis.'' Not reported on the transcript is the examiner's pointing to and even placing his (her) finger on the vulva or even inside the doll's vagina. The child, then, takes her finger and puts it in the same orifice. It's almost like playing the game, ``First you put your finger in, then I'll put my finger in.'' Children are the world's greatest imitators. They adhere slavishly to the dictum, ``If you're going to do it, I'm going to do it too.'' And that's a good enough reason for a three-year-old. And in this insidious way, yet another sex-abuse accusation becomes ``validated,'' with absolutely no evidence in the transcript of what has gone on.

Readers of transcripts are not generally aware of the significant amount of activity that is taking place with the dolls while the evaluator and child are talking. This is especially the case with regard to the dressing and undressing of the dolls. I have seen many tapes in which the child, after a short period of ``sex play'' with the dolls, wants to dress the dolls. Typically, the examiner will find a wide variety of excuses for discouraging the child from doing so, even to the point of physically preventing the child from putting on the doll's clothing. Obviously, the evaluator wants to prolong the period of exposure of the child to the naked dolls. And even when the child is engaged in activities with absolutely no sexual import (usually only a small fragment of these interviews are devoted to nonsexual material), the naked dolls are prominently present within the child's visual field and, obviously, serve as a reminder that they can once again be played with. Generally, it is only at the end of the session, just before it is time to leave, that the dolls are dressed. But even during the dressing phase, in which both the examiner and the child are engaged, there is visualization and touching of the sexual parts.


The primary purpose of this chapter has been to clarify the use of the term leading question and to suggest that it be subcategorized for the purposes of understanding better the various levels of leading questions and to improve communication among those who are using this term. This purpose is better served if evaluators are appreciative of the various degrees of leading stimuli (anatomical dolls being one example); again, subcategorization into four classes is warranted. Last, I introduce the term leading gesture, to refer to a maneuver that frequently operates simultaneously with leading stimuli and leading questions. My hope is that evaluators will be more sensitive to the importance of this factor in sex-abuse evaluations because it has not been given the attention it deserves. For this factor, also, four categories of classification are warranted.