Scottish jury research: findings from a mock jury study

5 How do jurors understand the not proven verdict?

Key findings

• The meaning and consequences of the not proven verdict were rarely discussed at any length in deliberations, even in those juries where that verdict was returned.
• Where the not proven verdict was discussed, there was inconsistency in understanding of its meaning and confusion over its effect. In particular, jurors were not always clear how it differed (if at all) from a not guilty verdict.
• The judge's direction that "not guilty and not proven have the same effect, acquittal, which means that the accused cannot be tried again for the same offence" does appear to increase juror understanding. Jurors in two-verdict juries, who did not receive this direction, were more likely than juries in three-verdict juries to think that the accused can be retried if the verdict is not proven.
• The idea that the not proven verdict means the accused is guilty, but that guilt has not been proven to the necessary standard for conviction, arose frequently (though usually briefly) during deliberations. Jurors also expressed the view that there is a lingering stigma attached to a verdict of not proven.
• Related to this, jurors choosing the not proven verdict tended to base their decision on a belief that that the evidence did not prove guilt beyond reasonable doubt, or on the difficulty in choosing between two competing accounts.
• Jurors choosing the not guilty verdict (where both acquittal verdicts were available), in contrast, tended to base this on a belief that the accused was innocent, or some aspect of the complainer's or witness' evidence that suggested that they were not giving a truthful account.
• More jurors thought that a verdict of not proven should be returned when jurors need to compromise to reach a verdict than believed a not guilty verdict should be used in that situation. However, there was also a view that the not proven verdict was a "cop out".

5.1 Introduction

As outlined in Chapter 1, Scottish juries are unique in that they have two verdicts of acquittal open to them: not guilty and not proven. The legal consequences of the two verdicts are exactly the same - the accused is cleared of the charges and cannot normally be re-prosecuted for the same offence.^[86] The not proven verdict's existence in Scots law has been described as a "historical accident".^[87] There were no set forms of verdict used by early juries: their role was simply to decide on the guilt or innocence of the accused. The role of the jury was altered in the early 17th century by a change in procedure whereby juries ceased to declare accused persons guilty or innocent, and instead returned 'special verdicts' considering whether individual factual allegations were proven or not proven. The decision on the guilt or innocence of the accused was then taken by the judge presiding over the case. In 1728, a landmark legal case (the trial of Carnegie of Finhaven) re-established the jury's right to return a verdict of not guilty, rather than leaving that decision to the judge. By the 19th century, lawyers had come to view the old 'special verdicts' as irrelevant. 'Not proven', however, had become something of a legal fixture, and juries continued to use it alongside 'guilty' and 'not guilty'. This was not in its original meaning, where 'not proven' referred to a failure to prove individual facts, but as one of two acquittal verdicts that both meant a failure to prove guilt. The not proven verdict as it is used today is not defined in statute or case law. It is simply one of two possible acquittal verdicts and the standard text on Scottish criminal procedure states that juries should not be told anything about its meaning.^[88]

This chapter presents findings on how jurors understand the not proven verdict and why they might choose it over the other verdicts available to them. It draws on data from two sources: individual questionnaires completed by jurors before and after their deliberations, and the jury deliberation transcripts and videos.

Of the 64 mock juries in our study, half (32) had two verdicts open to them (that is, they could only return a verdict of guilty or not guilty) and half were able to choose between three verdicts (guilty, not guilty or not proven). In his final directions to jurors, the judge in the trial films told them which verdicts were available to them. Jurors in two-verdict juries were told simply that there were two verdicts open to them - guilty and not guilty. In three-verdict juries, jurors were given additional direction on the not proven verdict, based on the guidance given to Scottish judges for directing juries in real trials, as follows:

"Finally, I need to tell you that there are three verdicts you can return on this charge: not guilty, not proven, or guilty. Not guilty and not proven have the same effect, acquittal, which means that the accused cannot be tried again for the same offence." ^[89]

Aside from this direction, jurors were told nothing about the not proven verdict. If jurors did ask questions about it, the researchers told them that they were unable to provide any further guidance. The not proven verdict was not mentioned at all in the questionnaire that jurors completed prior to deliberating, except that jurors were asked to state what they thought the verdict should be (so jurors in three-verdict juries had the option of choosing not proven). After deliberating, however, jurors in both two-verdict and three-verdict juries were asked a number of questions about their understanding of the not proven verdict. A questionnaire with the same questions about the not proven verdict was also completed by 'spare' jurors (those who watched the trial videos but who were not needed for the deliberations).^[90] Subsequent references to post-deliberation questionnaire data therefore includes the responses of these 'spare' jurors as well as those who actually deliberated.

5.2 Extent of discussion of not proven during deliberations

The research team coded the level of discussion of not proven's meaning or consequences in each jury deliberation, recording whether this was discussed "in some detail", only "minimal references" were made, or it was not discussed at all.

Across the study it was rare for any very lengthy discussion of the not proven verdict to take place, even in juries where that verdict was ultimately returned. The code "in some detail" therefore covered anything other than the most minimal discussion.

There were no statistically significant differences in the level of discussion of not proven between 12 and 15-person juries, or between juries asked to reach a unanimous verdict and those required to reach a simple majority (see Annex J, Table J.9). Juries in rape trials were marginally more likely than their assault trial counterparts to discuss not proven "in some detail", although they were no more likely to return a verdict of not proven at the end of the deliberation process.^[91]

Unsurprisingly, however, the not proven verdict was much more likely to be discussed "in some detail" by juries that had it available to them. In 11 of the 32 three-verdict juries, the not proven verdict was discussed "in some detail". Most of those juries (nine out of the 11) were juries in which a not proven verdict was returned. However, in five of the 32 three-verdict juries there was no discussion at all of the meaning or consequences of the not proven verdict, despite the fact that, in four out of these five, a not proven verdict was ultimately returned. Moreover, in two out of these five juries, several jurors switched their verdict from either guilty or not guilty to not proven during the course of the deliberations.^[92] There was no obvious pattern in terms of trial type, majority required, or length of deliberation to the five juries where the not proven verdict was not discussed at all.^[93]

It is notable that there was also at least some discussion of the not proven verdict in half (16 of 32) of juries in the two-verdict condition, despite there having been no mention of it in the judge's directions or pre-deliberation questionnaire. This indicates that there is wider awareness of the not proven verdict among the general public. The nature of this discussion was either simply to point out that a not proven verdict would normally be available in a Scottish criminal trial, or to lament the fact that it was not available to them - usually because jurors were finding it difficult to decide on a verdict. For example:

"I don't think there is enough evidence against [the accused], and that's it in a nutshell, it's quite clear. I would like to have a not proven verdict or something like that but can only go for not guilty."

(Assault trial / two-verdict / 15-person / unanimity)

5.3 Self-assessed understanding of not proven

After returning a verdict, all jurors were asked to assess how well they felt they understood the not proven verdict (on a scale of 1 to 7, where 1 was "I do not understand the not proven verdict at all" and 7 was "I fully understand the not proven verdict"). Of course, self-assessed understanding does not necessarily equate to actual understanding.^[94] The ways in which jurors actually understood the not proven verdict are explored in the next section of this chapter (5.4). However, 51% of all jurors felt that they 'fully understood' the not proven verdict.

Self-assessed understanding of not proven was higher in three-verdict juries (58% indicated that they fully understood the verdict, compared to 45% in two-verdict juries).^[95] There were no statistically significant differences in average levels of self-assessed understanding of not proven by gender, age, education level, trial type, jury size or majority required.

During deliberations, jurors sometimes expressed uncertainty or confusion about the not proven verdict. In nine of the 32 three-verdict juries, jurors either stated that they did not understand the not proven verdict or asked other jurors (or in one case, the researcher) what it meant. The nature of their uncertainty varied. In one jury, it was centred on whether not proven was actually a verdict of acquittal:

"Well, acquitting is not 'not proven', that's not guilty."

"Is that not guilty?"

"No, that's not proven, you must acquit..So, basically not saying he was guilty."

"So, you add the not proven and not guilty together."

"I don't understand that."

(Assault trial / three-verdict / 12-person / simple majority)

In two other juries, jurors expressed uncertainty about whether not proven and not guilty are the same thing, or whether not guilty differs because it requires proof of innocence.^[96] In another, a juror asked whether not proven would automatically be the verdict if the jury was split 50/50 between acquittal and conviction. Thus, while the judge's instructions appear to go some way to improving juror understanding of the not proven verdict, a lack of clarity about its precise meaning and use persisted among some jurors.

It should be stressed, though, that while there was some uncertainty over the meaning of the not proven verdict, jurors relatively rarely expressed beliefs about the verdict that were definitively incorrect. This is in part because, as noted earlier, the not proven verdict does not have a specific definition beyond it being one of two verdicts of acquittal. This leaves room for a number of different understandings of its meaning and purpose, which are explored below.

5.4 Specific understandings of the not proven verdict and its consequences

After deliberating, jurors were asked a number of questions about the not proven and not guilty verdicts as part of the post-deliberation questionnaire. These were designed to explore their understanding of the differences between the two acquittal verdicts, and why jurors might choose one over another. This section reports on findings from these questions and on the various ways in which the not proven verdict was discussed during deliberations.

5.4.1 "Guilty, but you can't prove it"

The idea that the not proven verdict should be used if jurors suspect the accused is guilty, but feel that this has not been proved beyond reasonable doubt, was the most prominent theme in both the deliberations and questionnaire responses.

A majority (70%) thought that if a jury thinks the accused is guilty, but do not think the evidence proves it beyond reasonable doubt, they should return a verdict of not proven. Just 7% said that it does not matter which of not proven or not guilty is returned, and 12% that the jury should return a verdict of not guilty (Figure 5.1). The view that not proven is the most appropriate verdict if jurors think the accused is guilty but the standard of proof has not been met was more common among: rape trial jurors than assault trial jurors (75% compared to 64%); jurors who had the not proven verdict available during their deliberations compared with those asked to choose between guilty and not guilty (77% compared 63%); and jurors asked to reach a simple majority compared with those asked to reach a unanimous verdict (73% compared to 66%).^[97]

Figure 5.1: Juror views on the appropriate verdict if they think the accused is guilty but the evidence does not prove it beyond reasonable doubt

Base: All jurors (n = 969)

The view that not proven should be used when the jury thinks the accused is guilty but that the evidence does not meet the required standard of proof was also voiced during deliberations - 31 such statements were made across 14 of the 32 three-verdict juries.^[98] For example:

First juror: "Not guilty is completely different to not proven."

Second juror: "Not proven is guilty, but you can't prove it."

First juror: "Not officially."

(Rape trial / three-verdict / 12-person / simple majority)

5.4.2 "Not proven means you could still be brought back to trial"

As noted above (section 5.1), jurors in the three-verdict juries were told by the judge that "not guilty and not proven have the same effect, acquittal, which means that the accused cannot be tried again for the same offence". Jurors in the two-verdict juries did not receive this direction.

In fact, the legal position regarding retrial following an acquittal verdict is now slightly more complex than this direction implies. Since the Double Jeopardy (Scotland) Act 2011 came into force, the prosecution can apply for permission to re-prosecute following an acquittal, although only in limited circumstances (primarily where either the acquitted person has subsequently admitted to committing the offence or, in serious cases, where fresh evidence has arisen that substantially strengthens the case against them).^[99] Analysis of the deliberations indicated that a small number of jurors in this study were aware of these provisions. The important point, however, is that such an application for re-prosecution can be made regardless of whether the verdict is not guilty or not proven. There is thus no difference between the two acquittal verdicts in relation to the possibility of retrial.

Our mock jurors were much more likely to think that a retrial was possible after a not proven verdict than after a not guilty one - 41% thought it was definitely or probably true that 'the accused can be tried again' after a not proven verdict compared with 23% who thought the same applied after a not guilty verdict (Figure 5.2).

Figure 5.2: Jurors' understanding of the not proven and not guilty verdicts' effects (% saying statements definitely or probably true)

Base: All jurors (n = 969)

The judge's direction does, however, appear to improve jurors' understanding of this issue. Jurors in two-verdict juries (who were not directed on this point by the judge) were much more likely than those in three-verdict juries (who were directed) to think that the position on retrial was different for each acquittal verdict (Figure 5.3). This is consistent with the findings of Hope and others, in which jurors who had been directed on the not proven verdict were significantly less likely to believe that the accused could be re-tried for the same offence following a not proven verdict.^[100] However, even in three-verdict juries, jurors were still more likely to say someone can be retried after a not proven verdict (26%) than after a not guilty verdict (20%) (Figure 5.3). Some confusion about the possibility of a retrial with each verdict appears, therefore, to persist. There were no statistically significant differences in views about the possibility of retrial by jury size or by majority required.

Figure 5.3: Juror beliefs about the possibility of retrial after (a) not proven and (b) not guilty verdicts, by whether they were asked to choose between two or three verdicts

Bases: All jurors - two-verdict jurors = 484, three-verdict jurors = 485

Whether the accused could be retried following a not proven verdict was also raised repeatedly during deliberations. In total, 21 statements on this topic were made across 12 of the 32 three-verdict juries: 11 expressing the view that this was possible, and 10 expressing the view that it was not.^[101] Jurors' claims that an accused person can be retried following a not proven verdict were not always clearly incorrect, given the possibility of retrial under the Double Jeopardy (Scotland) Act 2011. However, in five cases (in five different juries), jurors wrongly claimed that retrial is possible following a not proven verdict, but not following a verdict of not guilty:

"Not proven means that you could still be brought back to trial again. Not guilty means you can't."

(Rape trial / three-verdict / 12-person / unanimity)

Only one of these five statements was corrected by another juror, but this did not seem to be effective, as the juror who made the original statement went on to express the incorrect belief again.

All of these statements about not proven and the possibility of retrial occurred in the context of general discussions about not proven's meaning and how it might differ from not guilty. There were no examples of jurors referring explicitly to the possibility of retrial as a factor that led them to choose one verdict over the other, although it is, of course, possible that this influenced their decision.

5.4.3 "The middle ground"

When asked which verdict should be used 'when the jurors need to compromise to decide on a verdict' significantly more jurors selected not proven than not guilty (31% compared to 17%, respectively) (Figure 5.2, above).

The notion of not proven as a 'compromise verdict' also arose during deliberations, with statements to this effect made in five of the three-verdict juries.^[102] All five of these juries ultimately returned a not proven verdict. This view of the not proven verdict was sometimes linked with an expression of relief that the verdict was available as a way of ending deliberations:

" … if you didn't have the not proven verdict and you either had to find him guilty or not guilty then ..."

"We would be here all week."

(Rape trial / three-verdict / 12-person / unanimity)

However, there were also examples of jurors describing the not proven verdict as a "cop-out", suggesting that it excused the jury from deliberating more fully or from making a more difficult decision:

" … it's a bit of a cop out, rather than people really considering things really, really, carefully".

(Rape trial / two-verdict / 12-person / unanimity)

"… most people go for the middle ground because it's the easiest option and I think the danger of a situation like this, is that because you're not 100 per cent sure, which none of us are, you're just thinking it's safer just to say not proven …"

(Rape trial / three-verdict / 15-person / simple majority)

5.4.4 "If you've got a not proven for rape on your criminal record you're not working anywhere"

A verdict of not proven is a verdict of acquittal. It would not, therefore, form part of the accused's record of criminal convictions held by the court (which is given to the judge to take into account in sentencing if the accused is ever convicted of a subsequent offence).

Despite a not proven verdict not being a conviction, it is possible that information about it could be disclosed to future employers as part of a Protecting Vulnerable Groups (PVG) background check.^[103] This could occur if the fact of the not proven verdict has been recorded by the police and is considered by them to be relevant to the purpose of a PVG check under the relevant legislation (section 47 of the Protection of Vulnerable Groups (Scotland) Act 2007). The treatment of the not proven verdict in this way is no different from the not guilty verdict - either could potentially be disclosed as part of a PVG check in certain circumstances.

When asked directly (in the post-deliberation questionnaire), mock jurors did not appear to draw any significant distinction between not proven and not guilty in terms of their implications for the accused's criminal record. There was no statistically significant difference in the (relatively low) proportions of jurors who thought that the accused would get a criminal record following a not proven compared to a not guilty verdict (11% for not proven vs 8% for not guilty - see Figure 5.2).

This finding sits slightly at odds, however, with analysis of the content of jury deliberations. Although statements about the effect of a not proven verdict on the accused's criminal record were relatively rare, there was some evidence of misunderstanding. Ten statements were made on this topic across five of the 32 three-verdict juries.^[104] Six out of 10 statements implied that a not proven verdict would result in a criminal record, and four that it would not. Of the six statements indicating that a not proven verdict would appear on the accused's criminal record, four could be classed as definitely incorrect - they wrongly stated that if the accused was tried for a different criminal offence in the future, the jury would be told about the previous not proven verdict.^[105] The two other statements referred to the perceived effect a not proven verdict might have on subsequent employment. One made brief reference to possible disclosure of a not proven verdict to potential employers, while the other was a longer passage of discussion, starting with the following claim:

"If you've got a not proven for rape on your criminal record, you're not working anywhere, or you have no life anyway, so you're going to suffer anyway."

(Rape trial / three-verdict / 15-person / unanimity)

This was followed by a debate about whether someone who had received a not proven verdict would have to disclose this to future employers. No definitive conclusion was reached and the discussion moved on, with no further reference to this issue.

This pattern was common across other juries that discussed the perceived implications of not proven for the accused's 'record' - regardless of whether incorrect statements were corrected, the issue did not appear to have any bearing on subsequent discussions. Overall then, while there is evidence that some jurors may mistakenly believe a not proven verdict has different implications to not guilty for the accused's record, this did not appear to be an explicit driver for choosing one acquittal verdict over another (although again, it is impossible to know if individual jurors may have been influenced by this belief but did not voice that during deliberations).

5.4.5 "Not proven means the jury feel that they cannot prove the person is innocent"

The idea that a not proven verdict should be used when the accused has not 'proved his innocence' arose in 10 of the 32 three-verdict juries, with 15 statements made to this effect.^[106] There is, of course, no requirement to 'prove' innocence in legal proceedings, and all juries were directed on this by the trial judge: ^[107]

"I now deal with some fundamental principles of law that apply in every case. The first is this. Throughout the trial every accused is presumed innocent unless proved guilty. The accused is not required to prove his innocence. Secondly, it's for the Crown to prove the guilt of the accused on the charge he faces. If that's not done an acquittal must result. The Crown have the burden of proving guilt. Thirdly, the Crown must establish guilt beyond reasonable doubt."

Nonetheless, it was apparent that some jurors felt not proven should be used when innocence has not been 'proved':

"[Not proven] means that the jury feel that they cannot prove that the person is innocent, it means that they don't think...they think that there has possibly been a rape, but they don't feel that the evidence is strong enough."

(Rape trial / three-verdict / 12-person / simple majority)

In some instances, jurors distinguished between the not proven and not guilty verdicts in precisely these terms (that is, that not guilty indicates that the accused has proven their innocence, whereas not proven indicates that they have not):

" … the difference between not proven and not guilty is not guilty is you believe there is evidence to prove that he is not guilty."

"But, the outcome is the same."

"The outcome from our perspective is the same."

(Assault trial / three-verdict / 15-person / simple majority)

In the jury quoted above, this assertion led to a short debate, with another juror disagreeing that the not proven and not guilty verdicts differed in this way. Despite this, the juror who expressed the initial view then repeated it, stating that:

"Not guilty is there is evidence to prove that he did not do what they were accused of. Guilty is there is evidence that tells you. Not proven tells you there is no evidence."

(Assault trial / three-verdict /15-person / simple majority)

Discussion then moved on and the point was not discussed again (the jury ultimately returned a majority not proven verdict). In two other juries where this view was expressed, other jurors also interjected to disagree and challenge (in both cases referring to the judge's direction that the accused is innocent until proven guilty), but the point was not discussed further in either case, which makes it difficult to assess what impact such views had on the verdict these juries returned. However, there were examples where a belief that the accused had not proved their innocence appeared to be one reason why individual jurors favoured a not proven verdict over one of not guilty:

"I couldn't say he was innocent and let it go, but I couldn't say he done it, it was right in the centre for me."

(Rape trial / three-verdict / 15-person / simple majority)

5.4.6 "You've got a black mark against you"

A final theme that emerged during deliberations was the perception that there is an element of stigma attached to a not proven verdict. Seven statements were made to this effect across six of the 32 three-verdict juries (four in assault trial juries; three in rape trial juries).^[108] It was suggested that a not proven verdict means "you've got a black mark against you", or that "doubt" would exist in people's minds about whether the accused was guilty. This perceived 'stigma' was linked with the view (discussed in 5.4.1) that juries should return a not proven verdict when they think the accused is guilty, but the evidence does not prove this beyond reasonable doubt.

"You walk away innocent, but everybody knows."

(Rape trial / three-verdict / 15-person / simple majority)

In all these cases, it was clear that jurors understood the legal position (that a verdict of not proven is a verdict of acquittal), but that despite this, they felt there would be a lingering stigma attached to such a verdict.

Beliefs about the perceived 'stigma' of a not proven verdict were explicitly cited by three jurors as a reason why they favoured a particular acquittal verdict. All three were rape trial jurors. Two jurors (in different juries) stated that they supported a not guilty verdict over one of not proven because, as one put it, they were "not prepared to imply someone's committed rape". Conversely, the third juror stated that their choice of not proven was intended to send a message to the accused that they doubted his story. These views did not appear to influence other jurors - indeed two of the jurors expressing opposing views were on the same jury and neither explicitly stated that they had changed their mind as a result of the other person's opinion - but it cannot be ruled out that other jurors were influenced by these statements about stigma without expressly saying so.

5.5 Reasons for choosing one acquittal verdict over the other

Establishing exactly why jurors chose a particular verdict was often challenging. Jurors did not always give clear or illuminating reasons for favouring a particular verdict (either in their questionnaire or during deliberations), and those in three-verdict juries almost never explained exactly why they had selected one acquittal verdict over the other. However, comparison of jurors' stated reasons for choosing (a) not proven and (b) not guilty (drawing on both questionnaire data^[109] and jurors' deliberations) suggests that there are some general differences in reasons for choosing the two acquittal verdicts.

In summary, jurors choosing the not guilty verdict (where both acquittal verdicts were available) tended to justify this in one of two ways:

• The first was that they believed the accused was innocent. This contrasts with jurors who chose the not proven verdict, who tended to justify this on the basis that guilt had not been proved beyond reasonable doubt (sometimes accompanied by a belief that the accused was probably guilty, and sometimes by a belief that there was not enough evidence to form a view either way).
• The second was that they believed one or more of the prosecution witnesses were lying. Again, this contrasts with those who favoured not proven, who tended to justify this on the basis that they found it difficult to decide between the truthfulness of the different witnesses' accounts.

A more detailed comparison of reasons for choosing not guilty and not proven in each trial shows that jurors who favoured not proven were: ^[110]

• More likely to refer to the difficulty of choosing between the accounts given by the complainer and the accused (28% of rape trial jurors who thought the verdict should be not proven, compared to 12% of jurors who thought the verdict should be not guilty; the equivalent figures for the assault trial were 17% vs 6%).
• For the assault trial only, significantly more likely to refer in general terms to whether there was enough evidence to convict, or to additional evidence they felt was needed (for example, 60% of assault trial jurors who thought the verdict should be not proven cited insufficient evidence to convict vs. 41% of assault trial jurors who thought the verdict should be not guilty).
• For the rape trial only, significantly more likely to refer specifically to whether the case had been proved 'beyond reasonable doubt' (34% of rape trial jurors who thought the verdict should be not proven cited this vs. 19% of rape trial jurors who thought the verdict should be not guilty).
• In the rape trial only, jurors who favoured not proven were less likely to refer to specific elements of the evidence, and tended instead towards more general statements, such as 'not enough evidence to convict' (42% of jurors who thought the verdict should be not proven referred to specific elements of the evidence, compared with 65% of jurors who thought the verdict should be not guilty).

(See Annex J, Tables J.11 - J.13 for more detail).

These findings are consistent with discussions during the deliberations, where - among those jurors who expressed a reason for their choice of verdict - there was clear variation in the kinds of reasons cited by those who chose not proven and those who chose not guilty. Stated reasons for favouring not proven often reflected jurors' different understandings of not proven discussed above - for example, choosing it because they were not convinced of the accused's innocence. Jurors also noted difficulty choosing between the accounts given by the accused and the complainer as a reason for choosing not proven (in both assault and rape trials):

"Well, the thing is if you had two children and you had a falling out and you weren't there, which of them would you believe? You know, if you had two kids and they had a disagreement about something would you say you're guilty, or would you say I'm not really sure and not proven? I would say, well I wasn't really there so I can't decide which of you, I'm not 100 per cent sure. That's why I would say as well not proven, because I wasn't actually there, I didn't see what happened."

(Assault trial / three-verdict / 12-person / unanimity)

"I think it's very difficult with this because it's only one against one, there is no other witnesses. And I mean it's kind of a ... quite a hard thing to say which one is innocent or not, I'm finding it difficult. One minute we're looking at I was with him, and the next minute I was with her, and I'm finding it difficult … I would say not proven, because I'm not sure."

(Rape trial / three-verdict / 15-person / unanimity)

Other jurors linked their choice of a not proven verdict to the corroboration rule. As noted earlier (section 4.3.2), all juries were directed that the accused cannot be convicted on one source of evidence alone (in line with the way juries would be directed in a real Scottish criminal trial). This direction was picked up among jurors choosing a not proven verdict, particularly in the rape trial. For example:

"Yes, I'm going not proven, because I think that [the complainer] is credible and reliable and I believe her testimony, but as the judge said, we need to have corroboration and I don't think it was beyond reasonable doubt corroboration."

(Rape trial / three-verdict / 15-person / unanimity)

A final related theme, which again was most prominent in the rape trial, was jurors justifying their choice of not proven on the basis that they personally thought the accused was guilty but were not sure of this beyond reasonable doubt. For example:

"I think he was guilty, but I would probably have to go for a not proven because I think the fact that they couldn't prove beyond whatever reasonable doubt that she got the bruises from an attack means that really reluctantly I would probably have to go not proven. Although on the basis of her testimony and his, I definitely believe her."

(Rape trial / three-verdict / 15-person / simple majority)

Again, this is consistent with the finding, discussed above (section 5.4.1), that 70% of all jurors believed not proven is the most appropriate verdict when jurors think the accused is guilty but that this has not been proved beyond reasonable doubt.

There was evidence of unease among some rape trial jurors about choosing the not proven verdict when they believed that the accused was probably guilty. For example, one juror who thought not proven was the correct verdict nonetheless spoke of the accused "walking away" if he was in fact guilty and found this "difficult". Another, who also supported a not proven verdict on the basis that the case had not been proved beyond reasonable doubt, nonetheless felt that this verdict was "unfair" to the complainer and could be seen as "saying that we're sanctioning rape".

5.5.1 Reasons for switching between acquittal verdicts

69 of the 431 jurors (16%) deliberating in three-verdict juries switched between the two acquittal verdicts over the course of deliberations.^[111] This switch was most commonly from not guilty to not proven (12% of three-verdict jurors switched in this direction, with 4% changing from not proven to not guilty).

The scope for analysis of reasons for switching between acquittal verdicts is limited by the low numbers - of the 69 jurors who switched, only 48 gave a reason for doing so in their post-deliberation questionnaire.^[112] Moreover, the reasons given were not always particularly illuminating - the most common reason was simply that they had been persuaded by the discussion (67% of jurors who switched between acquittal verdicts and gave a reason for having done so). Other reasons for switching between acquittal verdicts generally reflected the themes already discussed: jurors' views by the end of deliberation on whether there was enough evidence or proof in general (33%); the perceived reliability and credibility of the complainer (23%) and witness (19%); and difficulties in choosing between inconsistent accounts or evidence (17%).

There were only five examples of jurors giving an explicit account during deliberations of their reasons for switching between the two acquittal verdicts. Again, their explanations reflected jurors' reasons for favouring not proven or not guilty, discussed above - for example, becoming convinced of the accused's innocence (as a reason for switching from not proven to not guilty), or difficulties choosing between the accounts of the accused and the complainer (as a reason for switching in the opposite direction).