back to criminal law back to home

Summing up - a judge's perspective

[2006] Crim L R 817

One juror has described the judge's summing up as the climax of the trial. He felt like a horse that had been indifferently ridden and even mistreated for the past four months, suddenly in the hands of an expert horseman. [1]

What, though, is the point of the judge's summing up? To direct the jury on the relevant law, to summarise the prosecution and defence cases and to review the evidence. Also, importantly, to give the jury a structure which they can use in their decision making process - what Lord Devlin described as "an agenda for their discussions". [2] These answers may seem obvious, but sometimes it is important to go back to first principles, to strip apart and question the daily things that we know best, about which we are most familiar. The summing up is often the most important step that a judge takes in a trial, prior to sentence. It is also, if done well, one of the more difficult. Communication between judges and advocates is comparatively easy. We speak the same language, use the same jargon. Communication between judges and lay people can be more difficult. 

Legal directions

Traditionally, the summing up starts with the judge's directions on the law. But do jurors remember the legal directions? Do they understand those parts which they do remember? We assume that they do, otherwise there would be no point in giving them, but we do not know.Since 1981, it has been a contempt of court "to obtain, disclose or solicit any particulars of statements made, opinions expressed, arguments advanced or votes cast by members of a jury in the course of their deliberations." [3] This has largely prevented academic research into the way in which juries work and the effectiveness (or otherwise) of judges' summings-up. However, research by the Centre for Criminology at Middlesex University found that more than half the jurors in their study did not fully understand what was happening in court and many were uncertain about how to ask a judge a question, or even whether they were allowed to take notes. Jurors said that the main impediment to understanding proceedings was the use of legal terminology.[4] Research in New Zealand suggests that misunderstandings of a legal nature are common among individual jurors and that in the main comprehension and recall problems are associated with the oral nature of the criminal trial process. 62% of jurors said that it would have been helpful if they had been provided with key legal directions in writing. The New Zealand research also indicated that much time and effort was wasted by juries trying to recall the directions given by the judge.[5]

Some forty years ago, when research into juries' deliberations was not so constrained, W. R. Cornish, professor of English Law at the London School of Economics, spoke to numerous jurors about their time at court. In his classic book, The Jury, after pointing out that there were no written directives which the jury could take away, he wrote

There is a natural tendency to disregard what is said about things which the jury cannot understand; nice distinctions over the precise meaning of a rule of law, or the judge's assessment of expert evidence, may thus pass into oblivion.[6]

He also carried out research with mock juries. That showed how little effect judges' legal instructions on matters such as the standard of proof, corroboration and recent complaint had on verdicts and "how much the care lavished on their form by appellate courts has been lawyers'shibboleth". He was sceptical about proposals to reform the content of directions which "can be appreciated only by those who have training or experience enough to do so." [7]

Judges' legal directions ought to be short, clear and simple. As recently as 1979, Lord Devlin was able to write that "the judge should briefly explain the law".[8] In an ideal world, directions would be given in language easily understood by all lay people and comprise no more than the burden and standard of proof and the legal ingredients of the offence charged.[9] They might even be as simple as "What the prosecution witnesses described was an assault occasioning actual bodily harm. What the defendant described was lawful self defence. If you are sure that the prosecution evidence is true, the defendant is guilty. If you think that what the defendant said may be true, he is not guilty."

But life in the Crown Court is not so simple. The legal directions which have to be given have become increasingly complex. Sir Robin Auld has described the legal element of summing up as "a long and burdensome journey for judge and jury alike."[10] Guidance from the Court of Appeal and the ever increasing number of criminal statutes has added to the length and intricacy of directions on the law. Sir Robin Auld referred to them as "highly technical and detailed propositions of law. Many are prolix and complicated, often subject to qualifications and in some instances barely comprehensible to criminal practitioners never mind those who have never heard them before."[11] The Judicial Studies Board model forms of legal directions are sophisticated and fair, but they are also lengthy, and, at times, complex. The current JSB Crown Court Bench Book[12] containing specimen directions is some ninety-four pages long. Besides covering the ingredients of the offence, a summing up may have to include directions on inferences to be drawn from one defendant's failure to answer questions in the police station,[13] the effect of another defendant's lies in the police station[14], the way in which a jury may use a defendant's bad character[15] (or good character) and the need for caution when relying upon evidence of visual identification[16]. If the JSB specimen directions are followed, those four directions alone amount to over three pages of A4 single spaced typing. 


Although there is no English or Welsh research into how much jurors remember or understand, there is research by psychologists into memory, recall and understanding. In 1956, George A. Miller concluded that "the span of immediate memory impose[s] severe limitations on the amount of information that we are able to receive, process and remember."[17] His research suggested that there was a limit of seven chunks of information (plus or minus two) which could be retained. More recently, and more significantly, Robert H Margolis of the University of Minnesota , has written about how much patients remember from what they have been told by doctors and other health care professionals. According to Margolis, studies indicate that about 50% of information provided is retained. Between 40 and 80% may be forgotten immediately. In one study, patients could not recall 68% of the diagnoses given to them about their own health in a medical visit. Of the information that is recalled, about half is remembered incorrectly. So about half is forgotten immediately and half of what is remembered is wrong. In other words, patients remember correctly only about a quarter of what they are told. Intelligence has not been shown to affect the proportion of information retained, but elderly patients do tend to remember less than younger patients. Information presented in simple, easy-to-understand format is remembered better than information presented in a more complex manner. The more information presented, the lower the proportion that is recalled by the patient. Written material, when used appropriately during a consultation, can enhance recall of information.[18]

If patients forget half of their own diagnoses and get half of what they do remember wrong, what chance is there of jurors remembering legal directions? And what is the effect on jurors' verdicts, if they only remember correctly about a quarter of the legal directions?

Written directions

In many American states the charge to the jury (including the legal directions) is given in written form. In New Zealand jury research has perhaps contributed towards greater use by judges of written material when directing juries.[19] Increasingly, judges in Canadian provinces are leaving the legal aspects of the charge in written form with the jury. Cornish wrote that the American practice of giving written instructions to the jury "should be given consideration in England." [20]. Writing in 1989, Professor Edward Griew of Nottingham University also called for judges to reduce the jury's task to a series of written questions in every case.[21]

There is little authority on the giving of written directions in English and Welsh courts. In R v McKechnie[22] the judge, towards the end of his summing up, indicated that he would provide the jury with a written list of questions to assist their deliberations. In the absence of the jury, counsel suggested alterations to the questions which he accepted. The judge gave the written material to the jury, finished his summing up and identified the agreed alterations, but did not take the jury through the questions or relate them to his summing up. The written questions were not always in the same form and did not follow the same structure as the treatment of the issues adopted by the judge in his summing up. In the Court of Appeal, Auld J said

It is sometimes useful in a complicated case for a judge to supplement his summing up by a written list of directions or questions to a jury. That would be so where the questions individually are complicated and/or where there are many of them and it is essential that the jury should answer them in a particular order. However, if a judge considers it necessary to supply the jury with written directions, he should prepare them and submit them to counsel for their consideration in good time before they begin their closing speeches.

The importance of giving a copy of any written material to counsel in advance was reiterated in R v Wright,[23] where the judge prepared directions which he dictated to the jury. The Court of Appeal held that he should have discussed them first with counsel at the end of the evidence so that they could be agreed and counsel could deal with them in closing speeches.

In R v Lawson (Raymond)[24] a jury requested written directions on the law. The judge declined their request. The Court of Appeal held that the judge was fully entitled to decline to direct the jury otherwise than orally.

Giving written directions goes against the tradition of orality in British justice, but that is a tradition which has already been eroded in the civil courts where skeleton arguments from advocates are often expected - indeed many skeletons are so fleshed out that they are positively obese. Most judges' experience shows that it easier to assimilate complex legal submissions if the oral advocacy is reinforced by written skeletons.

Zander and Henderson's Crown Court Study revealed that in 1993 judges gave juries written questions to simplify their task in only 2% of cases.[25] In more complex criminal trials judges now give written materials to juries more frequently. Generally they take the form of a series of questions or steps for the jury to work through. It is though still rare for judges to give typed versions of all their legal directions. Why is this? If giving skeleton arguments to judges is helpful, isn't giving written directions to jurors, who often have no previous experience of court hearings and so are unfamiliar with legal thought processes, likely to be of far more assistance in assimilating their contents?

Over recent months, I have adopted the practice of giving written directions to jurors in all cases, however straight forward. Before closing speeches, I give the advocates a typed draft of my legal directions and discuss them with them. If necessary I make alterations. I sum up in the normal way, but when I conclude the legal directions, I indicate to the jury that when I have finished summing up the facts, I will give them a typed version of the legal directions. When I do this, saying that I hope that the written material will be helpful, jurors frequently nod. The typed legal directions are identical to the oral directions and comprise both the ingredients of the offence and other legal directions (such as the Turnbull identification direction, section 34 inferences, good or bad character etc). Generally, the typed version is handed to the jury by the usher at the very end of the summing up because I want the jury to concentrate on what I say throughout the summing up, rather than be tempted to leaf through the written material. However, if it is appropriate to give the jury a series of questions or steps to work through, I include them at the end of the typed version of the legal directions and the usher gives out the combined typed legal directions and questions immediately before I go through the questions with the jury and move on to sum up the evidence.

Handing out written directions seems to have almost eliminated requests from juries for reminders or further guidance on the law. Juries also seem to be reaching verdicts more quickly. There has also been an increase in the proportion of convictions. This is borne out by an analysis of verdicts in my court over the last two years.[26] In those cases where the jury were not given a typed version of the legal directions, 44% of defendants were found guilty, 47% were found not guilty and the jury was discharged because there was no prospect of them reaching a verdict in respect of 9% of defendants. In cases where juries were given a typed version of the legal directions, 60% of defendants were found guilty and 37% were acquitted. Perhaps most significantly, to date, only one jury which had typed directions has failed to reach a verdict. 

This sample is so small that it has little or no statistical validity. Little reliance can be placed on the result of cases in one judge's court. However, these percentages underline the need for proper research into the effect of giving juries written directions. Perhaps though in the absence of such research, a few tentative theories may be advanced from this experience. It would be logical to assume that written directions are likely to provide a more effective structure to a jury's decision making process. They may well act as a more effective agenda, to use Lord Devlin's words, which can be used as a tool by those jurors who are keener to follow the judge's legal directions. Some of the more complex pro-prosecution directions (e.g. inferences which may be drawn from defendants' silence during interview and bad character) may have little impact when read out in the course of a judge's summing up, but may be of significance when juries have a typed version in front of them, in black and white in the jury room. Written directions may also mean that, if jurors do not have to argue about their recollections of what the judge said about the law, they can have more clearly focussed discussions on the evidence.

Uncertainty and confusion about legal directions are likely to lead to doubt and so acquittals which do not reflect the evidence. As William Young, a judge of the High Court of New Zealand, wrote, after reviewing the New Zealand jury research, a "wayward" verdict is perhaps more likely to result from the jury misunderstanding either the law or the evidence than for any other reason. So, "The incidence of perverse verdicts is likely to be reduced if judges can be more effective in ensuring that juries decide cases in accordance with the law and the facts[27] Giving written directions is clearly one way of doing this and of increasing the likelihood that the guilty are convicted and the innocent are acquitted.

The principal objection to giving written directions is the additional work load that it places on judges. Judges write their summings-up before or after the normal 10 a.m. to 4.30 p.m. court day. They are part of the substantial unseen work done by judges outside the court room.However for those judges who already prepare their legal directions on computers, tidying them up and correcting typing errors so that they can be given to juries is not a great deal of extra work. For those judges who do not prepare their summings-up in this way, there is a strong business case for HM Court Service providing the appropriate IT support to enable legal directions given orally to be converted into a typed version, either by use of voice recognition software or a programme such as LiveNote. It would be well worth the expenditure if it reduced the number of hung juries and resulting re-trials.

continue to second part of article back to criminal law back to home 

[1] Trevor Grove The Juryman's Tale p174. The expert horseman was HHJ Simon Goldstein.

[2] Patrick Devlin Trial by Jury p116.

[3] Contempt of Court Act 1981 s8.

[4] and The Guardian Only 43% of jurors grasp all the details 27 January 2004. It appears that the sample was not asked about judges' summings up. See too Zander and Henderson, The Crown Court Study, Royal Commission on Criminal Justice Research Study no 19, HMSO, 1993,

[5] William Young Summing-up to Juries in Criminal Cases  What Jury Research says about Current Rules and Practice [2003] Crim L R 665.

[6] Cornish The Jury p125.

[7] The LSE Jury Project Juries and the Rules of Evidence [1973] Crim L R 208 at 222.

[8] Patrick Devlin, The Judge, p100, emphasis added.

[9] R v McVey [1988] Crim L R 127.

[10] Review of the Criminal Courts of England and Wales , p533. Sir Robin's solution, that jurors should have a written case and issues summary supplemented by a list of written questions, has not been taken up.

[11] Review of the Criminal Courts of England and Wales , p534.

[12] Described by Sir Robin Auld as a crib for judges. Available at

[13] Criminal Justice and Public Order Act 1994 s34.

[14] R v Lucas 73 Cr App R 159.

[15] Criminal Justice Act 2003 ss98 to 112.

[16] R v Turnbull 63 Cr App R 132.

[17] George A. Miller, The Magical Number Seven, Plus of Minus Two: Some Limits on Our Capacity for Processing Information, The Psychological Review 1956 vol 63 pp 81-97.

[18] Robert H. Margolis Boosting Memory with Informational Counselling The ASHA Leader August 3, 2004, p10 available at and In One Ear and Out the Other  What Patients Remember available at

[19] William Young Summing-up to Juries in Criminal Cases  What Jury Research says about Current Rules and Practice [2003] Crim L R 665 at p 669. For jury experiments in New York, see Sand and Reiss A Report on Seven Experiments Conducted by District Judges in the Second Circuit (1985) 60 N. Y. U. L. Rev. 423, pp453 -460.

[20] Cornish, The Jury p126.

[21] Edward Griew Summing-up the Law [1989] Crim L R 768.

[22] (1992) 94 Cr App R 51. For the position in the USA, see e.g. Hopt v People 104 U. S. 631 and Haupt v United States 330 U. S. 631 and Scharzer Reforming Jury Trials 132 F. R. D. 575 at p. 584 but cf People v Owens 69 N.Y2d 585 (in New York consent of counsel required in criminal cases). In New York , after research by the Unified Court System's Jury Trial Project, the Committee on Jury instructions is now recommending that judges routinely supply deliberating juries with a copy of the final charge - see Krauss, Jury Trial Innovations in New York State N. Y. State BJ 22.

[23] [2000] Crim L R 510.

[24] [1998] Crim L R 883.

[25] Zander and Henderson, The Crown Court Study, Royal Commission on Criminal Justice Research Study no 19, HMSO, 1993, p216.

[26] Sample of all cases at Harrow Crown Court in which I summed up to a jury between June 2004 and June 2006. Of 43 defendants in cases with no written directions, 19 defendants found guilty on one or more counts, 20 defendants acquitted on all counts, 4 cases where jury discharged with no verdict. Of 30 defendants in cases with written directions, 18 defendants found guilty on one or more counts, 11 defendants acquitted on all counts and one jury discharged with no verdict.

[27] William Young Summing-up to Juries in Criminal Cases  What Jury Research says about Current Rules and Practice [2003] Crim L R 665.