‘[Expletive],that was confusing,wasn’t it?’ Defence lawyers’ and intermediaries’ assessment of the language used to question a child witness

Abstract

While language specialists and legal professionals have voiced concerns about the language used to question child witnesses in the Aotearoa/NZ courts, it is unclear whether both groups share a common understanding of what those language problems are. This study compares five Aotearoa/NZ defence lawyers’ and two England/Wales intermediaries’ perceptions of the developmental (in)appropriateness of the language used to question an 11-year-old witness, based on their assessment of the witness’ anonymised trial transcript. The comparison showed that both groups agreed on the categories of language features that might confuse children, however, intermediaries identified many more instances of problematic language within those categories than lawyers. Training on developmentally appropriate language and pre-trial preparation of questions would certainly help lawyers improve the comprehensibility of their questions. However, the implementation of a full intermediary scheme, such as that in England/Wales, probably offers the best prospects for a sustained sea change in questioning practices.

Keywords

advocacy children cross-examination developmentally appropriate language intermediaries

Introduction

…because it can’t be right that it’s the…proper function of the lawyer to create ambiguity. (Defence lawyer)

There has long been disquiet within adversarial jurisdictions internationally about the criminal courts’ treatment of child witnesses, with concerns raised about (inter alia) the developmental appropriateness of the language used to question them and the potential for poor questioning to diminish the quality of their evidence. Such concerns in England/Wales led to the development of the Registered Intermediary Scheme, where accredited intermediaries (most of whom are speech language therapists (SLTs)) are engaged to facilitate effective communication between vulnerable witnesses and the courts (Ministry of Justice (England/Wales), 2015). Their role includes assessing a witness’s communicative competencies, then advising the court on how best to question that witness (e.g., which language and other features the witness can or cannot handle), monitoring courtroom questioning and alerting the judge if a question is unlikely to elicit best evidence.¹ A recent experimental study shows the positive impact of intermediaries on the quality of children’s evidence,² while a series of qualitative studies have shown that the majority of the police, counsel and judges consulted were highly positive about intermediaries’ ability to facilitate the testimony of young and vulnerable witnesses (Hanna et al., 2010; Henderson, 2015; Plotnikoff and Woolfson, 2007, 2015; Victims’ Commissioner (England/Wales), 2018). In Aotearoa/New Zealand, s. 80 of the Evidence Act allows for intermediaries (‘communication assistants’) to similarly facilitate witnesses’ participation in proceedings. However, communication assistants are rarely engaged to assist in child witness trials, despite the fact that many questions posed by lawyers are likely to exceed children’s comprehension and diminish their ability to produce best evidence, particularly in cross-examination (Davies and Seymour, 1998; Hanna et al, 2012; Zajac and Cannan, 2009; Zajac, Gross and Hayne, 2003).

Lawyers and judges in Aotearoa/NZ have also voiced unease about the language used to cross-examine children. When asked which language features they considered problematic, the lawyers and judges were vague, citing ‘age inappropriate’ and ‘confusing’ questions, ‘playing on semantics’ and poor phrasing (Davies et al, 2011). They acknowledged that courtroom examinations can easily go awry because they cannot always recognise when questions are beyond a child’s comprehension. They lamented their lack of knowledge about, and available training on, how to match questions to young witnesses’ competencies. While intermediaries in England/Wales can help counsel adapt their language for the witness, the rarity of communication assistants in the Aotearoa/NZ courts means that the vast majority of counsel and judges must simply do the best they can unaided.

Language specialists will inevitably be better equipped to recognise problematic language than most lawyers. What is less clear is where the overlaps and gaps lie between language specialists’ and lawyers’ perceptions of what constitutes developmentally inappropriate language: When these groups say there are problems with the way children are questioned, are they talking about the same thing? Judging by transcripts of children’s courtroom questioning in Aotearoa/NZ (Davies and Seymour, 1998; Hanna et al, 2012; Zajac and Cannan, 2009; Zajac, Gross and Hayne, 2003), one would conclude that the gaps are many and wide, replete as those transcripts are with features that can confuse children.³ Some instances no doubt arise because lawyers overestimate children’s competencies; other instances may be intentional (Henderson, 2002). However, many problems likely arise because inculcation into the formal language and jargon of the courts can make it difficult for an experienced lawyer (or judge) to shift gear sufficiently to avoid legalese and complex language when questioning a child—to block the complex syntax that characterises the unplanned speech of the highly literate (Miller, 2006) and the set phrases characteristic of courtroom speech (e.g., ‘Is it not the case that…?’). Lawyers frequently formulate questions on the spot in response to the testimony elicited, making it even harder for them to attend to the wording of a question as well as its tactical function—while keeping an eye on the jury’s reactions. Hence, extrapolating lawyers’ understanding of developmentally appropriate language from performance data (i.e., transcripts) is unlikely to produce reliable results. Nor did asking legal professionals to cite examples of problematic language elicit useful information (Davies et al., 2011), perhaps because they did not have the analytic tools to name those features. Conversely, asking lawyers to point to problems in the transcript of (unknown) colleagues’ questioning may show us more clearly where agreement and disagreement lies on what constitutes developmentally appropriate language.

The current study

The current study compares a small group of Aotearoa/NZ defence lawyers’ and England/Wales intermediaries’ perceptions of the developmental appropriateness of the language used to question a young witness, based on their assessment of an anonymised transcript of the witness’s courtroom questioning. The aim is to ascertain where the gaps and overlaps lie in the groups’ assessment of what constitutes developmentally appropriate language and provide a forum for both groups’ perspectives to be heard to inform thinking on how to improve questioning practices.

Defence lawyers, rather than prosecutors, were chosen for two reasons. First, most children who testify in the Aotearoa/NZ criminal courts do so as complainants in sexual offence cases (Hanna et al., 2010). While the questions posed by prosecutors (and judges) are certainly not above reproach, it is the language of cross-examination which has attracted particular attention in the literature. Secondly, cross-examination typically comprises the bulk of courtroom questioning because children’s evidence-in-chief is usually replaced by their digitally recorded police interviews (EVIs) (Hanna et al., 2010); prosecutors’ questions are largely restricted to supplementary questions after the EVI and re-examination, which is normally brief. Registered Intermediaries from the UK were chosen as the language specialists because of their unique combination of language expertise and courtroom experience.

The current study partly replicates Krähenbühl’s (2011) where intermediaries and lawyers assessed the mock examinations of a 6-year-old. The main differences are that the current study uses the actual transcript of an 11-year-old’s courtroom examinations (with minor modifications) as few local cases involve such young witnesses (Hanna et al., 2010). It also included participant interviews to further explore assessments of the language used in the transcript and views on questioning children. Finally, this paper reports on participants’ perceptions regarding word choice and syntactic complexity, rather than on question types and other features of the cross-examination (to be reported separately).

Method

Participants, recruitment, procedure

The participants comprised a convenience sample of five experienced Auckland defence lawyers (one female, four male) and two English intermediaries (both female). The lawyers all qualified in the 1980s or 1990s. Two are now QCs and at least three have taught advocacy (defence lawyers are identified herein as D1 to D5). The intermediaries are qualified speech language therapists with decades of clinical practice and extensive intermediary experience. At the time of the interviews (2014), one had assisted with around 78 trials and the other with 30 involving children and other vulnerable witnesses (intermediaries are identified as Int1 or Int2). Recruitment was via personal contacts and snowballing; participation was on the basis of informed, voluntary consent.

All participants received an anonymised transcript of a child’s courtroom questioning, along with associated documents, also anonymised.⁴ They were asked to identify any questions which may be developmentally inappropriate for the 11-year-old witness, to explain why the question was inappropriate, and where possible to rephrase the question appropriately. Participants received the following non-exhaustive definition of ‘developmentally inappropriate’:

questions or language which could confuse the child (e.g., difficult vocabulary, concepts, grammar; poorly worded questions);

questions which are phrased in a way that could elicit inaccurate or ambiguous responses from the child; and

any other feature of the questioning/utterance which could impede the child’s ability to provide accurate evidence (e.g., sequencing of questions; leading questions).

After participants had returned the marked-up transcript, they took part in individual semi-structured interviews to probe their comments and explore their views on the language of cross-examination, in practice (e.g., from a strategic perspective) and in theory (e.g., what constitutes good advocacy), given that the two can be in conflict. Interviews lasted between 1 and 2 hours and were conducted in 2014 and 2015. All participants were paid for their time at standard professional rates. The study was approved by AUT’s ethics committee.

The transcript

The transcript comprised the courtroom questioning of an 11-year-old complainant in a sexual offence trial held in Aotearoa/NZ in 2012. As the child’s evidence-in-chief was given via her EVI, substantive courtroom questioning comprised supplementary questions following the EVI, cross-examination and re-examination. All court documents were anonymised by the Auckland District Court before release to the researchers and participants. Minor modifications were made to the transcript to increase the range and/or number of possible problems; the modifications involved features attested in transcripts of Aotearoa/NZ children’s courtroom examinations from 2008 (see Appendix). The modified transcript comprised 287 questions/utterances by the judge (n = 37), prosecutor (n = 58) and defence (n = 192), of which 220 were substantive.⁵

Findings

Participants were deemed to have identified developmentally inappropriate language if they specifically criticised a feature in their transcript analysis or interview and/or rephrased a question to remove a feature.

Word choice

Legalese

Intermediaries and lawyers were uniformly critical of the legalese phrases ‘I put it to you’⁶ and ‘I suggest to you’.⁷ Lawyers labelled them ‘very old-fashioned’ (D5), ‘formalistic’ (D4) and ‘antiquated’ (D3, D4); all agreed they are unnecessary:

…those sorts of expressions are just legal verbiage which add nothing, except that they generally suggest that defence counsel are girding their loins ‘cos they know they’re asking crunchy questions that they’re likely to get bad answers to. (D1)

There was less agreement among lawyers about the appropriateness of other common legal expressions identified by one or both intermediaries as developmentally inappropriate, such as ‘is it your evidence that’ (identified by D2, D4); ‘defendant’ (identified by D1, D3, D4);⁸ ‘(give) evidence’⁹ (identified by D1, D2); ‘interview’¹⁰ (identified by D1, D2, D4); ‘His Honour’ (identified by none). Formulaic expressions using more everyday language were also highlighted as inappropriate by intermediaries and lawyers, such as ‘Do you say…?’¹¹ (Int1, Int2, D4) and ‘Is that the position?’¹² (Int1,¹³ D1, D3, D4).

Formal language

All participants identified the word ‘protrude’¹⁴ as developmentally inappropriate; both intermediaries and three lawyers edited out the word ‘routine’¹⁵ (D1, D4, D5). Beyond that, there was little consensus between lawyers on what constituted formal vocabulary/expressions that might be poorly understood by an 11-year-old. For example, one of the five lawyers removed the words ‘recognise’¹⁶ and ‘on this occasion’, noting that the latter seemed a ‘formal way to phrase the question’ (D4); one removed ‘refer to’¹⁷ (D1) and ‘to be mistaken’¹⁸ (D4); two removed ‘to be positioned’¹⁹ (D1, D4);²⁰ and two questioned the use of the word ‘(in)appropriate’ (D2, D3). All of these terms were either highlighted or removed by both intermediaries,²¹ along with other formal words that elicited no comment from lawyers (e.g., ‘mention’,²² ‘prepare to’²³). One lawyer (D2) removed the words ‘panic’²⁴ and ‘gentleman’, describing the latter as ‘not age-appropriate’; intermediaries did not comment on the former, but removed ‘gentleman’ in their rephrases.

Using the child’s (and other witnesses’) terms

Lawyers were keenly aware of the need to adopt the child’s terms in their own questions (e.g., terms for body parts and people’s names), drawing attention to instances where counsel in the transcript failed to do so. Two lawyers (D1, D2) also pointed to instances where counsel failed to faithfully report other people’s speech. For example, the defendant claimed to have used the terms ‘erect’²⁵ and ‘testicles’ when talking to the child, however, defence asked her whether the defendant had mentioned ‘erections’ and ‘testes’. As D1 noted, the child’s response may have differed had defence used the correct terms. Intermediaries likewise highlighted the need to maintain consistency in this regard, but also pointed to more subtle instances. In the following exchange to check camera alignment at the beginning of the child’s court appearance, the judge misreported the child’s response in line (b):

1. (a) Judge: What am I doing?

(b) Child: Swinging your glasses.

As Int1 pointed out, this may be a very minor point, but had potential to disconcert the child from the outset, ‘making her feel she didn’t quite get it right and unsettle her for future responses…’

Clarity of reference and meaning

Intermediaries and lawyers frequently drew attention to cases where confusion could arise due to unclear reference. For example, to more clearly identify the intended referent, both intermediaries and two lawyers (D2, D4) replaced third person personal pronouns (e.g., ‘he’, ‘she’, ‘him’) with full noun phrases (e.g., ‘Dan’). Intermediaries did so in 35 questions and lawyers in seven. Both intermediaries and four lawyers (D1-D4) replaced common nouns (‘the lady’) with proper nouns (‘Jan’); similarly, both intermediaries and one lawyer (D1) replaced a locative ‘there’ with ‘in the room’. Intermediaries also replaced pronouns referring to actions and states with full descriptions to improve clarity, which lawyers tended not to do (D1 and D5 did so in one case). For example, the pronoun ‘that’ in line (c) refers to an action—but it could have meant getting into bed or touching the penis. The intermediary’s rephrase in line (d) removed all doubt:

2. (a) Judge: Defence: And when you got into bed with Dad, did you touch his penis?

(b) Child: Yes.

(d) Rephrase (of c): Did you touch his penis when he was asleep? (Int1)

Intermediaries and lawyers were particularly concerned about unclear temporal references. For example, the use of ‘earlier’, ‘after’, ‘before’ elicited criticism where the point of comparison was unclear (e.g., earlier than what?) as did the phrase ‘that time’ where there were two possible candidates for the temporal referent. Participants also highlighted words which the child was probably familiar with, but whose meaning in the context may have been unclear. For example, in his introductory comments, the judge said, ‘…if you’re not happy or you feel uncomfortable, you just tell me and I’ll fix it, okay?’ Both intermediaries removed the reference to ‘uncomfortable’. As Int1 explained, the word probably means something to the child like ‘feeling itchy and fidgety’, rather than unhappy. Polysemous words like ‘right’ and ‘wrong’²⁶ were identified as potentially problematic by one intermediary (e.g. ‘right’ can mean ‘correct/true’ or ‘morally good’), while lawyers questioned whether the child understood the significance of ‘just’ below (the lawyer presumably intended it to mean ‘only’):

3. Defence: He just told you that it was inappropriate to talk about sex…

However, only intermediaries picked up on the potential for confusion arising from figurative language (‘keep it to yourself’) and the difficulty some children around the age of the witness have distinguishing between ‘ask’ and ‘tell’.²⁷

Syntactic complexity

Voice

There were nine passive clauses in the transcript, all but two appearing in substantive questions. Seven were specifically identified by at least one of the intermediaries as passives and potentially problematic and/or removed in rephrased versions. They noted that children’s ability to process passives can vary:

…[passives are] liable to potentially cause more processing difficulties, particularly if the child’s anxious…given lots of time and lots of thinking they’d be able to cope with it but in the live cross-examination they wouldn’t. (Int1)

Interviewer: Are passives still problematic for an 11 year old? Intermediary: Not necessarily, but I think we just need to default to assuming that they could be. I think you can’t guarantee you’re going to get an accurate answer, so why use them? (Int2)

No lawyer specifically remarked on the use of the passive voice in the questions, although D3 rephrased a passive clause into the active voice in the marked-up transcript (again, without making a comment about the passive). Another question involving a passive clause (4 below) was identified by three lawyers as confusing, although they attributed the complexity to the coordination of ‘asleep’ and ‘waking up’ rather than to the passive voice. When asked during the interview how he would rephrase this question, D2 spontaneously produced versions using the active voice. It is not clear whether this was a conscious choice or simply reflects the unmarked (i.e., most basic and common) status of the active voice.

4. Defence: And the only time that Dad’s penis was touched was when he was asleep or waking up? Child: Yes.

Rephrase: ‘Would you agree, the only time that you touched Dad’s penis, he was asleep?’ or ‘The only time you touched Dad’s penis, he was just waking up’ [D2]

Negation

Fifty-one of the utterances posed to the child in the modified transcript contained negatives,²⁸ including 43 in substantive questions, comprising 19.5 per cent of all substantive questions.²⁹ In 17 of these 43 cases, the negative appeared in reported speech (e.g., ‘And did Dad tell you not to panic about…?’). These negatives may have been difficult to avoid without employing an open question (e.g., ‘What did Dad say?’), which defence can be loath to do. Of the 26 remaining substantive questions, intermediaries rephrased 18 to remove all negatives, or at least one of multiple negatives, or specifically identified a negative as problematic; lawyers did so in relation to three. Negatives in non-substantive questions elicited less comment, perhaps because they were less likely to affect the quality of the child’s testimony.

However, both intermediaries and most lawyers were particularly attuned to instances where a child’s unelaborated ‘yes’ or ‘no’ response to a negative question was ambiguous and required clarification. In the examples below, the responses appeared inconsistent with the child’s earlier testimony and now supported the defence case theory. The response to (5) was clarified by the lawyer; in the other cases, he swiftly moved on to a new topic.

5. Defence: And you’re quite sure that he wasn’t asleep when you touched his penis?

Child: No.

6. Defence: And on that morning he did not ask you to keep a secret, did he?

Child: No.

7. Defence: And that’s what he was doing, wasn’t it, not telling you to keep a secret?

Child: Oh, yes.

During interviews, one lawyer explained how he would have clarified the child’s response to (6) above:

…Normally what I would say…’And on that morning he did not ask you to keep a secret, did he?’ ‘No.’ ‘So you’re agreeing with what I’m saying?’ ‘Yes.’ And sometimes that’ll do it. So, ‘Did you just agree with me?’ ‘Yes I did.’ (D2)

However, this approach would itself be problematic. First, the question intended to clarify is a biased, declarative question (‘So you’re agreeing with what I’m saying?’) which expects a ‘yes’ response (Huddleston and Pullum, 2002). Secondly, it would not uncover any miscommunication occasioned by the negative ‘not’ in the first question, given that even single negatives can cause comprehension problems (Perera, 1984). The intermediaries’ rephrases, on the other hand, removed the original negative, hence even a bare ‘yes’ or ‘no’ response would be unambiguous:

8. On that morning did he tell you to keep a secret?…Are you sure? (Int1)

9. On that morning of [specifies day] did he ask you to keep a secret? (Int2)

Multiple, complex and lengthy questions

Lawyers and intermediaries identified 10 substantive questions containing more than one request for information or confirmation, for which different answers were possible (including one ‘planted’ multiple question).

10. Defence: [Did Dad sit on the side of the bed with you]_Q1 and [talk about sex]_Q2?

Child: Yes.

11. Defence: [I’m suggesting that you’re quite wrong]_Q1 and [that there were no (sperm) come out of Dad’s penis that morning]_Q2.

Child: No. Some did.

While lawyers picked up on multiple questions appearing in lengthy utterances, only intermediaries picked up on less obvious examples appearing in shorter questions:

12. Defence: Did you grab his penis_[Q1] and squeeze it_[Q2]?

13. Defence: And did Dad put on shorts_[Q1] and a t-shirt_[Q2]?

All lawyers, and intermediaries to a much greater extent, identified a host of unnecessarily long and/or complex questions posed by the judge, prosecutor and defence. Participants rephrased these to decrease their complexity, for example by (singly or in combination) removing verbiage (14), simplifying the syntax (e.g., removing cleft constructions³⁰ (15) and avoiding multiple clauses (16)), turning long questions into two shorter ones (17), and postposing subordinate clauses to avoid frontloading (18) (see Table 1).³¹

Table 1.

Participant rephrases of long and complex questions.

	Original	Rephrases
14	Judge: Do you promise me still you’re going to tell the truth when this lady and then following that a man is going to ask you some questions?	Do you promise to tell the truth when you answer questions? (Int1) Do you promise me still you’re going to tell the truth when you answer questions this afternoon? (D1)
15	Judge: Anne, what I want you to do is to promise me that when you’re talking to everyone here today that at all the time, all times, you’ll tell the truth.	Anne, I want you to promise to tell the truth today. Will you tell the truth? (Int1) Anne, I want you to tell the truth today. I want you to say only what really happened. (Int2)
16	Defence: Had he walked into the room earlier after having a shower, or did he leave to go and have a shower?	When did he have a shower? (Int1)
17	Defence: And when Dad talked to you about sex and touching, people not touching you, did he tell you that sex was not something you should talk about in [location]?	‘“Dad told you that sex [is] something you shouldn’t talk about in [location]”. If the timing matters a follow up question can be asked.’ (D5) Did dad tell you that sex was not something you should talk about in [location]? (Yes). Did he tell you that while you were sitting on the side of the bed? (Yes). (D1)
18	Defence: Did Dad, when you say that Dad said it was a secret, what were his words?	What did dad say when he said it was a secret? (Int1) What words did Dad say when he said it was a secret? (D4)

However, lawyers’ rephrases sometimes retained complex elements from the original question or introduced new ones. For example, in 19 (see Table 2), the lawyer divided the original question into two, but retained an avoidable negative. In 20, two lawyers shortened the original question, but still included an avoidable negative and added an interrogative tag. In 21, the lawyer felt the phrase ‘any worries’ was ‘too general’, however, the attempt to clarify resulted in a question frontloaded with three subordinate clauses, with another two elsewhere in the question. The meaning of ‘through the person from the Court’ may also be opaque to an 11-year-old.

Table 2.

Participant rephrases of questions.

	Original	Rephrases
19	Defence: And you’re quite sure that he wasn’t asleep when you touched his penis?	Wasn’t Dad asleep when you touched his penis?…Are you sure about that? (D4) When you touched his penis was Dad asleep?…Are you sure? (Int1)
20	Defence: Well I’m suggesting to you that at no time that morning did he take his shorts down?	Anne he never took his shorts down, did he? (D5) Dad didn’t pull his shorts down, did he? (D4) Are you sure Dad took his shorts down any time that morning? (Int2)
21	Judge: Anne, I just wanted to talk to you about a few things first of all. The first thing is if you’ve got any worries at any time you just tell me okay? Would you do that for me?	If you have any worries about what is happening in Court while you give your evidence you can always let me know through the person from the Court who is in the room with you. (D3)

The perceived benefits of clarity and confusion

During interviews, lawyers stressed the importance of using simple, non-’lawyerly’ language when questioning children, for the witness’ and jurors’ benefit:

…where [counsel] finishes the question, Is that the position?, well, that’s something that a lawyer says. Normal people don’t talk like that…I mean, more so a child, but to any witness when you say, Is that the position?,…to me it’s unnatural. (D1)

…it’s an oral situation where the jury have to just take it all in. They don’t have it written down. So if you’re not simple, they’ll miss your point. (D5)

They identified a range of strategic benefits arising from keeping questions simple. One noted that children can react badly to confrontational questions, such as those prefaced with the lawyerly clause ‘I put it to you’: ‘They think they’re being told off or that the adult disapproves of them’, which may lead to defiance or withdrawal (D5). Others argued that posing developmentally appropriate questions makes it easier for the jury to assess the child’s reliability as that judgement will not be mediated by concerns that the child was confused by the questions or how to answer them.

However, participants pointed out that clarity is not always what a cross-examining lawyer seeks. When the case is not strong—e.g., in the absence of corroborating evidence, which is typical of child sexual assault cases—a bit of confusion may be all that can be hoped for to create reasonable doubt:

…generally, as the defendant you won’t have that hard evidence, and therefore confusion is your friend in the sense that you can say to…the jury, ‘Well, look, she said this but who knows what she meant by that?’ The more you can point out confusion and inconsistency, that’s often the only thing you’ve got. (D1)

If you’ve got a really clear, cogent, strong defence, you don’t want ambiguity…Sometimes though ambiguity is a defence lawyer’s best friend because the onus is on the Crown…to prove beyond reasonable doubt and if you can create some doubt in the jury’s mind then you’re doing your job for your client. (D4)

However, manufacturing inconsistencies was perceived to be a step too far:

I draw the line at trying to trick the witness or manipulate them to the point where…they’re not actually giving evidence that they’re meaning. Or scaring them or bullying them. It doesn’t really work. (D5)

Language features can not only make a question difficult to understand, but also elicit ambiguous responses. For example, when negative declarative questions (with or without an interrogative tag) elicit an unelaborated ‘yes’ or ‘no’ response, as in (22) below, the meaning of that response can be unclear:

22. Defence: And on that morning he did not ask you to keep a secret, did he?

Child: No.

It is not at all obvious here whether the child meant, ‘No, he did not ask me to keep a secret’ or ‘No, you’re wrong, he did.’ Lawyers are duty bound not to mislead the courts, where ‘misleading’ may include ‘failing to correct a misleading impression’ (Robertson, Beck and Mount, 2014: xviii). Two lawyers were asked if counsel had a duty to clarify ambiguous responses such as that in 22. Both saw this as the judge’s responsibility, not counsel’s (D1, D3). Further, two lawyers argued that any decision by defence to resolve ambiguity or clarify a response would depend on whether it would help or hinder the case (D1, D4) and how important the point is (D4), leaving it to the judge to correct any potentially misleading impressions and/or to the prosecution to resolve them during re-examination. Of course, the ambiguity may not be immediately evident—in the real-time pressure of the courtroom, as D1 pointed out, counsel may interpret the response in one way, only for it to dawn later that the answer was open to interpretation. However, he also noted that, if there are question types like negative declaratives that can elicit ambiguous responses, he would have no difficulty with lawyers being instructed not to use them, ‘…because it can’t be right that it’s the…proper function of the lawyer to create ambiguity’ (D1). This is precisely what Int2 perceived to be happening in at least some courts in England/Wales: Lawyers are accepting intermediaries’ recommendations to avoid negative questions to the point that they are no longer being put to children.

Leaving aside any positive duty, lawyers pointed out that failing to clarify ambiguous responses can be strategically risky, even when the response appears to favour their case. First, if the jury recognised the ambiguity and/or the judge intervenes, it could put defence counsel in a poor light:

…it makes you look sneaky…as if you’re not getting to the truth, because you want to at least appear to be getting to the truth…It’s a terrible thing to say, isn’t it, but that’s the whole point. You need to be appearing to be a searchlight. (D2) (original emphasis)

…there’s also a good chance at this point that the judge is going to step in and do the clarification, and then you lose. Then the defence lawyer got the point but the judge cleared it up. The judge is on her side. So you don’t want the judge to be the clarifier. You want to be seen to be able to clarify yourself. (D2)

Secondly, if the prosecutor clarifies the response during re-examination, they will undo any apparent gain, leaving no further opportunity for defence to contextualise or deal with the (now) unfavourable evidence. However, leaving aside risks and benefits, one lawyer argued that taking advantage of language to create ambiguity is simply anathema to good advocacy:

…fundamentally I don’t think that the creation of confusion and ambiguity…is the most effective form of advocacy…even if you were defence in a sexual trial, if the only thing that you’ve got is things like negative declaratives which create ambiguous answers…you’re really absolutely scraping the barrel. (D1)

Nonetheless, one lawyer perceived that capitalising on ambiguous responses is sometimes a deliberate tactic:

…sometimes these tactics are actually deliberate and you’re counting on the prosecutor not being able to recover it. I personally think it’s a little dangerous but some people do adopt those tactics. (D3)

Despite the number of difficult questions in the transcript, at no point in the transcript did the judge intervene, not even to clarify ambiguous responses. Nor did the prosecutor object during cross-examination, although most of the unclear responses were cleared up during re-examination. If responsibility for ensuring fact-finders are not misled by ambiguous responses really does rest with the judge, then this judge failed in that duty. And many of his (and the prosecutor’s) own utterances were criticised by participants for being overly long, complicated, unclear and containing legal expressions. Nor did the child say she did not understand a question, even in the face of questions that the study participants themselves found confusing (and were not ‘plants’).³² Yet three lawyers (D2, D3, D4) commented on how well the child appeared to be answering the questions and concluded she must have been comprehending them. However, 72 per cent of the child’s responses to substantive questions comprised an unelaborated ‘yes’ or ‘no’,³³ which could easily have masked non-comprehension. Furthermore, the research literature is clear that children often answer questions they have not understood (Perry et al., 1995) and that child witnesses rarely tell the court when they do not understand (Hanna et al, 2010). As Int1 cautioned, giving a seemingly sensible answer does not mean the witness understood the question.

Pre-trial preparation and counsel training

In assessing the court transcript, intermediaries and lawyers identified many opportunities to improve the comprehensibility of the questions. However, D4 pointed out that it is one thing to identify problems when one is reading a transcript at leisure, but quite another to formulate appropriate questions in real time. One way to ease the real-time cognitive load could be to draft questions ahead of time to check for wording, sequencing and relevance before trial, which is now recommended practice in England/Wales. One lawyer in the current study stated that she always does this when preparing to examine children (D5). Another pre-prepared his questions at the beginning of his career, but no longer, to his regret (D2):³⁴

I wish I probably did, ‘cos I wouldn’t fall into any traps with anyone. No one would ever ask me to say the same question twice or appear to be confused. (D2)

The other three lawyers were opposed to drafting questions ahead of time, in particular for fear that this would impede the flexibility to respond to the emerging evidence:

…there’s always a danger that if you over-formulate your line of questioning you become a slave to that line of questioning as opposed to exploring avenues with the witness. (D4)

Given that lawyers typically approach cross-examination with a list of topics to be explored or points to be made, two (D3, D4) of those opposed to pre-preparing actual questions conceded that it would be possible to prepare some questions within those topics while still retaining flexibility to respond to the emerging testimony; one had recently done precisely that in a case involving multiple child witnesses where it was important with the first witness in particular to ‘establish the whole foundation of my case’ (D3).

Intermediaries confirmed that pre-preparing questions for child witnesses is becoming increasingly common in England/Wales, with some judges³⁵ instructing counsel to run their questions past the intermediary before trial—a practice endorsed by the England/Wales Court of Appeal.³⁶ Yet the approach still allows considerable flexibility:

They might have written down the core questions but then you just see on the next line it says, ‘Challenge such-and-such and such-and-such,’ giving themselves some flexibility to create a question, to suit what has just been said. (Int2)³⁷

However, while pre-preparing questions would undoubtedly remove some poor questions, it may not resolve all issues. Within this sample at least, there is a gap between lawyers’ and intermediaries’ understanding of what constitutes developmentally appropriate language, and not all of the lawyers’ rephrased questions were in fact developmentally appropriate. Training could help to bridge that gap, as the lawyers themselves noted. Having communication assistants/intermediaries monitor questions posed at court, as in England/Wales, would create an incentive to get it right:

If there was a person in court running a ruler over my questions and telling everybody that they weren’t entirely appropriate, I’d be doing a lot more preparation for my questions. I would make sure that my questions were unchallengeable, and I’d be saying, ‘Well, you ask it better.’ (D2)

Discussion and conclusion

…it’s not a case of saying to defence lawyers, ‘Don’t do this, because it’s mean to the children.’ It’s a question of saying, ‘Don’t do it, because it’s actually dangerous. It’s not the most effective way of cross-examining a child.’ (D5)

The stated aim of cross-examination is to challenge opposing witnesses’ evidence by, inter alia, revealing inconsistencies and improbabilities and ‘trying to make your client’s instructions and version [of events] possible’ (D5). However, taking advantage of children’s more limited communicative competencies to manufacture inconsistencies and improbabilities was viewed in a dim light by the participants, because it is not only poor advocacy, but strategically risky. Although lawyers in this study conceded that confusion can be helpful when the case is not strong, they also gave compelling strategic reasons for ensuring that children can understand the questions and that their responses are unambiguous, leaving the fact-finder in no doubt that children’s evidence is untainted by confusion and trickery.

The key question is whether lawyers are able to distinguish between questions which are and are not developmentally appropriate. The findings demonstrate that the intermediaries and lawyers in this study were largely in agreement on the categories of language features that might confuse children: word choice, unclear reference, negation, multiple questions and lengthy, complicated questions. The main difference lay in their assessment of what constituted confusing words, unclear reference and so on: intermediaries identified more of these features than lawyers. For example, intermediaries identified more potentially problematic formal or legal words and expressions than lawyers tended to; more types and examples of unclear reference; rephrased or highlighted more questions involving negation and the passive voice; and identified more subtle examples of multiple questions. They also highlighted word choices unremarked upon by lawyers, including figurative language and ‘ask v tell’. Intermediaries were also more successful than lawyers at consistently rephrasing complex questions to simple ones, without incorporating new complexities. These findings demonstrate the potential for lawyers to unwittingly confuse children. If jurors have similar blind spots, they too may not recognise when a child’s response could be contaminated by confusion.

A limitation of this study is that it involved a small sample, hence we cannot and do not claim that the findings are representative of the Aotearoa/NZ defence bar. However, the findings are consistent with those in Krähenbühl’s study (2011), where participants identified complex sentences, word choice, multiple questions and negation as potentially problematic. A gap between intermediaries’ and lawyers’ assessments within many of these categories was also evident. Furthermore, the problems counsel in this study had rephrasing complicated questions into simpler ones have been reported of lawyers in England/Wales (Plotnikoff and Woolfson, 2015) and lawyers and judges in Aotearoa/NZ (Hanna et al, 2012). Even within the small sample here, there was a great deal of variation between lawyers in terms of their perceptions of what constitutes appropriate language. We suspect the same is true of the Aotearoa/NZ defence bar more generally.³⁸

Despite this variation, as expected, the five lawyers individually and collectively missed many of the problems identified by the two intermediaries. Indeed, one lawyer reported that he initially could not see any problems with the questions at all; it was only after careful, close analysis that the problems became evident. Even then, the lawyers’ analysis of a written transcript is unlikely to reflect their ability to recognise issues arising in real-time oral questioning in an actual trial.

So how might legal professionals improve their knowledge of, and ability to produce, questions that children actually understand? Most lawyers agreed that pre-trial preparation of at least the core questions would help eliminate some inappropriate language. However, if the gaps in knowledge evidenced here are also true for other lawyers, there would still be a risk that not all problematic questions would be identified or appropriately rephrased. Training on developmentally appropriate language (including how to accommodate children with conditions that affect communication, such as autism spectrum disorder) would help ensure that lawyers’ assessments were based on evidence-based criteria. However, training may be resisted by those who believe that their questions are already perfectly fine (Davies et al, 2011) and/or, as some English lawyers believe, that talking to their own children and grandchildren provides all the training needed (Henderson, 2015). Judges too, as the individuals ultimately responsible for controlling courtroom questioning, require specialist training in developmentally appropriate language. But they will likely remain wary of intervening more often unless they are confident that the appellate courts would support a more robust approach (Henderson, 2016). Amending s. 85 of the Aotearoa/NZ Evidence Act to replace judges’ discretion to disallow unacceptable questions³⁹ with an obligation to do so might embolden some (see Victorian Law Reform Commission, 2016). If lawyers knew that judges would be policing questioning more vigorously, they might prepare questions more carefully to avoid embarrassing rebukes in front of the jury.

While training and pre-trial preparation would undoubtedly be beneficial, as participants themselves agreed, their effectiveness would depend on the skills and commitment of individual judges and counsel, which would likely vary. Furthermore, training could only give legal professionals generalised guidance on the communicative competencies of witnesses at different ages or with particular conditions, rather than individualised guidance on how to question a given child. Further, developing the skills to question very young witnesses and those with significant communication impairments would require extensive training and, even then, may still be beyond the skills of some counsel.

For these reasons, the England/Wales intermediary system seems to offer the best prospects for effecting a broad, sustained improvement in questioning practices, alongside counsel and judicial training and better pre-trial preparation. First, as small as the sample is, this study supports earlier findings that intermediaries bring far more expertise to the job than lawyers do. Using that expertise, intermediaries can complement any training lawyers receive by providing individualised guidance on questioning a particular vulnerable witness and giving judges better information on which to base decisions to intervene. Secondly, intermediaries in England/Wales are enabling children as young as 2 (Victims’ Commissioner (England/Wales), 2018) and severely incapacitated witnesses to be interviewed effectively, using skills and methods well beyond most lawyers’ competencies:⁴⁰ intermediaries’ ability to increase access to justice is justly celebrated. Finally, there is potential for ongoing collaboration between intermediaries and legal professionals to support, reinforce and extend the benefits of training. Lawyers and judges can learn incrementally from intermediaries, trial by trial increasing their awareness of, and skills in adapting to, young and vulnerable witnesses (Plotnikoff and Woolfson, 2015).

Aotearoa/NZ is slowly moving in the right direction, with a series of recent innovations which have potential to improve questioning practices. For example, in 2017, the District Courts initiated the Serious Sexual Violence Court pilot which (inter alia) required pilot judges to attend three days’ training with sessions by psychologists and SLTs on developmentally appropriate questioning. Furthermore, the courts and police occasionally engage SLTs as communication assistants. In the absence of local protocols, communication assistants model their practice on the England/Wales system. There have also been some short, ad hoc training seminars for lawyers. However, a sea change in questioning practices is unlikely until high-quality training is mandatory for all judges who hear child witnesses cases and, at the very least, available to counsel; until judges are emboldened to intervene more robustly to control questioning; and until well trained, appropriately qualified communication assistants become a more common sight in the nation’s courtrooms. Until that time, the gap between lawyers’ intentions and their practice is likely to remain wide and, at least by lawyers, widely under-appreciated.

Footnotes

Acknowledgements

The authors are immensely grateful to the participants for sharing their time and expertise, to AUT’s Faculty of Culture & Society Research Committee for funding this study, and to the Aotearoa/NZ District Courts for permission to access anonymised court documents. We are also greatly indebted to Dame Joyce Plotnikoff, Ali Gale, participants and the two anonymous reviewers for their insightful reviews of earlier drafts. Any remaining deficiencies remain the sole responsibility of the authors.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The researchers received funding from AUT University’s Faculty of Culture and Society Research Committee (see Acknowledgements above) but not from any other third party.

Notes

Appendix: Modifications made to the original transcript

Modifications

Words replaced
Original	Replacement	Replacement wording/feature also appeared elsewhere in original transcript
Poking	Protruding	No
Proper noun (name)	Defendant (x2)	No
Hard	Erect (reference to a penis)	No
Where were you	How were you positioned	No
I’m saying	I put it to you	Yes
Did she tell you if you didn’t tell her you wouldn’t be allowed to [specifies activity]	Did she tell you you can’t [specifies activity] unless you told her what happened?	No
Judge Smith	His Honour	No
Before	On a previous occasion	No
Are you saying…?	Is it your evidence that…?	No
Words added to existing questions
on this occasion		No
Modifications to questions
Combining two questions into one to produce a multiple question (x2)		Yes
Adding a question asking if the child was sure about her previous response (x2)		Yes

References

Astington

Olson

(1990) Metacognitive and metalinguistic language: Learning to talk about thought. Applied Psychology: An International Review 39(1): 77–87.

Carter

Bottoms

Levine

(1996) Linguistic and socioemotional influences on the accuracy of children’s reports. Law and Human Behaviour 20(3): 335–358.

Chomsky

(1969) The Acquisition of Syntax in Children from 5 to 10. Cambridge, MA: The MIT Press.

Collins

Harker

Antonopoulos

(2017) The impact of the registered intermediary on adults’ perceptions of child witnesses: Evidence from a mock cross examination. European Journal of Criminal Policy & Research 23(2): 211–225.

Crawford

Bull

(2006) Teenagers’ difficulties with key words regarding the criminal court process. Psychology, Crime & Law 12(6): 653–667.

Dando

Ormerod

Cooper

Marchant

Mattison

Milne

Bull

(2018) No evidence against sketch reinstatement of context, verbal labels or the use of Registered Intermediaries for children with autism spectrum disorder: Response to Henry et al. (2017). Journal of Autism and Developmental Disorders 48(7): 2593–2596.

Davies

Seymour

(1998) Questioning child complainants of sexual abuse: Analysis of criminal court transcripts in New Zealand. Psychiatry, Psychology and Law 5(1): 47–61.

Davies

Hanna

Henderson

. (2011) Questioning Child Witnesses: Exploring the Benefits and Risks of Intermediary Models. Auckland: Institute of Public Policy, AUT University.

Freshwater

Aldridge

(1994) The knowledge and fears about court of child witnesses, schoolchildren and adults. Child Abuse Review 3(3): 183–195.

10.

Hanna

Davies

Henderson

. (2010) Child Witnesses in the New Zealand Criminal Courts: A Review of Practice and Implications for Policy. Auckland: Institute of Public Policy.

11.

Hanna

Davies

Crothers

. (2012) Questioning child witnesses in New Zealand’s criminal justice system: Is cross-examination fair? Psychiatry, Psychology & Law 19(4): 530–546.

12.

Henderson

(2002) Persuading and controlling: The theory of cross-examination in relation to children. In Westcott

Davies

Bull

RHC

(eds). Children’s Testimony: A Handbook of Psychological Research and Forensic Practice. Chichester: John Wiley & Sons.

13.

Henderson

(2015) ‘A very valuable tool’: Judges, advocates and intermediaries discuss the intermediary system in England and Wales. International Journal of Evidence and Proof 19(3): 154–171.

14.

Henderson

(2016) Taking control of cross-examination: Judges, advocates, and intermediaries discuss judicial management of the cross-examination of vulnerable people. Criminal Law Review 3: 181–205.

15.

Henry

Crane

Nash

. (2017) Verbal, visual, and intermediary support for child witnesses with autism during investigative interviews. Journal of Autism and Developmental Disorders 47(8): 2348–2362.

16.

Huddleston

Pullum

(2002) The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press.

17.

Krähenbühl

(2011) Effective and appropriate communication with children in legal proceedings according to lawyers and intermediaries. Child Abuse Review 20(6): 407–420.

18.

Miller

(2006) Clause structure in spoken discourse. In Brown

(ed.) The Encyclopaedia of Language and Linguistics. 2nd ed. Boston, MA: Elsevier, 481–483.

19.

Ministry of Justice (England/Wales) (2015) The Registered Intermediary Procedural Guidance Manual (2015). London: Ministry of Justice. Available at: https://www.theadvocatesgateway.org/images/procedures/registered-intermediary-procedural-guidance-manual.pdf (accessed 22 August 2018).

20.

Nippold

(2007) Later Language Development: School-age Children, Adolescents, and Young Adults. 3rd ed. Austin, TX: PRO-ED Inc.

21.

Perera

(1984) Children’s Writing and Reading: Analysing Classroom Language. Oxford, England: Basil Blackwell Ltd.

22.

Perry

McAuliff

Tam

. (1995) When lawyers question children: Is justice served? Law and Human Behavior 19(6): 609–629.

23.

Plotnikoff

Park

(2018) In the blink of an eye. Archbold Review 1: 4–6.

24.

Plotnikoff

Woolfson

(2007) The Go-between: Evaluation of Intermediary Pathfinder Projects. London: Lexicon Limited. Available at: http://lexiconlimited.co.uk/wp-content/uploads/2018/03/Intermediaries_study_report.pdf (accessed 22 August 2018).

25.

Plotnikoff

Woolfson

(2015) Intermediaries in the Criminal Justice System: Improving Communication for Vulnerable Witnesses and Defendants. Bristol: Policy Press.

26.

Plotnikoff

Woolfson

(2017) Dispensing with the ‘safety net’: Is the intermediary really needed during cross-examination? Archbold Review 6: 6–9.

27.

Robertson

Beck

Mount

(eds). 2014. Introduction to Advocacy. Wellington: New Zealand Law Society CLE Ltd.

28.

Stolzenberg

McWilliams

Lyon

(2017) Ask versus tell: Potential confusion when child witnesses are questioned about conversations. Journal of Experimental Psychology: Applied 23(4): 447–459.

29.

Victims’ Commissioner (England/Wales) (2018) A Voice for the Voiceless: The Victims’ Commissioner’s Review into the Provision of Registered Intermediaries for Children and Vulnerable Victims and Witnesses. London: Office of the Victims’ Commissioner. Available at: https://victimscommissioner.org.uk/a-voice-for-the-voiceless/ (accessed 12 August 2018).

30.

Victorian Law Reform Commission (2016) Victims of Crime in the Criminal Trial Process: Report. Available at: www.lawreform.vic.gov.au/projects/victims-crime-criminal-trial-process/victims-crime-criminal-trial-process-pdf (accessed 12 August 2018).

31.

Walker

Kenniston

(2013) Handbook on Questioning Children: A Linguistic Perspective. 3rd ed. Washington, DC: ABA Centre on Children and the Law.

32.

Zajac

Cannan

(2009) Cross-examination of sexual assault complainants: A developmental comparison. Psychiatry, Psychology and Law 16(Suppl 1): S36–S54.

33.

Zajac

Gross

Hayne

(2003) Asked and answered: Questioning children in the courtroom. Psychiatry, Psychology and Law 10(1): 199–209.