Ch 16. Criminal justice

Almost everyone seems to believe in British justice. People argue about specific laws. Not everyone has confidence in the police. Some may think the judiciary is fuddy-duddy, or campaign for procedural reform. But hardly anyone appears to doubt that the natural consequence of crime should be to go before a court, and that in general the process is a good one. It is as though we imbibe confidence in British justice with our mothers’ milk. Yet in many respects we should be literally in contempt of court. It is neither as open nor just as it pretends, it is unnecessarily expensive and it is not very good at assessing evidence. These are in-house failings more interesting for the fact they are so readily overlooked than for the effect they have outside the system. More importantly from a crime prevention point of view, the criminal justice process is little more than a breakwater against the tidal rise and fall of crime. Most crime just swirls on by.

However provocative this may sound it is consistent with all the evidence. While specific criminal justice interventions may appear to have a short term effect there are no consistent effects over long periods of time or across different jurisdictions. The case against criminal justice as a crime prevention tool was summed up robustly in a submission to the Justice Select Committee in October 2013, “There is little empirical evidence that anything the Ministry of Justice has ever done since its inception in 2007 has had any impact on the commission of crime… the police have no idea what effective role they have had in reducing crime, no idea, based on evidence, about what works and little or no evidence to back any initiatives… Ignorance of, and lack of respect for evidence and the scientific method bedevils the whole area of crime policy” (Mine Technology, 31 October 2013 [] ).



In England and Wales, decisions on whether or not to charge a suspect are taken by the Crown Prosecution Service, which cherry-picks cases it thinks are ‘more likely than not’ to succeed. In other words, long before a public trial there is a private one and its outcome is highly subjective.

The CPS was formed in 1986 to take over prosecutions from the police and decide whether or not to proceed to court and what the charge should be. (Under French and similar jurisdictions a juge d’instruction or examining magistrate has overall charge of the investigation, and the Scottish procurator fiscal can supervise an inquiry, but cops still do the legwork.) The CPS has two key criteria, both of which require skilled judgement: whether or not there is enough evidence against the defendant and whether a prosecution is required in the public interest. The evidential benchmark is called the ‘Full Code Test’ and requires ‘a realistic prospect of conviction’ such that a guilty verdict is ‘more likely than not’. This is a lower burden of proof than ‘beyond reasonable doubt’ and the CPS maintains this is an ‘objective’ test. Self-evidently it is not – CPS lawyers are required to guess what the outcome of a trial would be. The CPS has tough financial targets too. In 2010 its budget of £712 million p.a. was cut by 25 per cent over the following four years.

CPS cherry-picking has perverse consequences for crime reduction as well as justice. Since the ‘Full Code Test’ is preoccupied with securing a conviction, a guilty verdict becomes an end in itself. Everything is reduced to a one-off event. Numerous crimes are ‘taken into consideration’, effectively reducing them to one. ‘Sample charges’ are proffered, which means just one charge is brought to represent many, often reducing a colourful pattern of offending to a monochrome event. Lawyers have told me, ‘you can’t put dozens of charges individually and expect a jury to make a decision… it would make trial impossible.’ But this simply restates the problem: that the courtroom defines our response to crime, imposing its rules and definitions even at the expense of disguising behaviour patterns of individual offenders and concealing trends in community safety. This goes to the heart of the tension between securing prosecutions and securing a safer society; the former focuses exclusively on discrete events that lead to individuals being acquitted or convicted; the latter seeks patterns and trends which it can disrupt to forestall crime in the future. (See Chapters 21 and 22.)





As one of their most senior lawyers told me, ‘We’re almost a judicial function at an earlier stage.’ And as a local CPS solicitor put it, ‘Prosecuting is quite a lonely job.’

Quotes are from contemporaneous notes and filmed interviews at CPS headquarters and CPS Thames Valley, 2009.




Despite big improvements by the CPS more than half of all cases fail to make the grade. They are what the Ministry of Justice calls ‘cracked’ or ‘ineffective’ trials, usually because of lack of courtrooms or judges, the unpreparedness of lawyers, or because witnesses don’t turn up or defendants fail to appear.

Figures have remained fairly constant for many years. In the second quarter of 2012, for example, ‘48 per cent of Crown Court were recorded as effective, 37 per cent were recorded as cracked and 15 per cent were recorded as ineffective’. Magistrate court figures were similar: ‘44 per cent were recorded as effective, 17 per cent were ineffective and 39 per cent were recorded as cracked’. ‘Court Statistics Quarterly, April to June 2012’, Ministry of Justice Statistics Bulletin, 27 September 2012, pp. 38–9.[]




They would not even be permitted to read scientific evidence from source materials – jurors have been imprisoned themselves for looking things up online.

In January 2012 a former university lecturer was jailed for six months – a sentence longer than is frequently imposed on seriously violent offenders – for carrying out online research about a criminal defendant while serving as a juror. The Lord Chief Justice said, ‘The damage to the administration of justice is obvious.’ Other jurors have gone much further. For example, in 18 August 2008 at Newcastle Crown Court, Judge David Hodson halted a homicide trial after six days of evidence because a diligent juror had visited the scene of the killing, taken measurements and then, with other jurors, sent the judge a list of questions which they felt had not been posed in court.




When Australian researchers were given unprecedented access to jurors immediately after trials had concluded, they found three-quarters of juries were not even agreed on whether the defendant had been acquitted or found guilty … in 40 per cent of cases at least one juror believed the judge had ordered an acquittal when no such direction had been made.

Judy Cashmore and Lily Trimboli, Child sexual assault trials: A survey of juror perceptions, Crime and Justice Bulletin, No. 102, September 2006, New South Wales Bureau of Crime Statistics & Research. In contrast to the UK ban on jury research the Attorney General of NSW can authorise juror surveys under section 68A of the 1977 Jury Act. Consent for this research was granted in order to assess the value of CCTV in cases of child sex abuse, and 277 jurors were interviewed from twenty-five juries in Sydney between 2004 and 2005. The shocking findings about jury confusion were thus incidental and were not included in the government press release of 18 December 2006. [The original paper is at$file/CJB102.pdf.]




In Britain juries are ‘remarkably effective decision-makers’1 even if the decision-making process is less than perfect. The former editor of The Times, Simon Jenkins, describes how a jury he served on ‘let a killer loose’ largely because ‘a couple of taxi drivers wanted to get back to work’.2 A large-scale simulation of trials in Britain found a majority of jurors failed to understand the judge’s directions and many of them defied instructions, including trawling the internet for evidence that had not been presented in the courtroom.3

1Cheryl Thomas, The Times, 25 February 2013, p. 21.

2 Simon Jenkins, The Guardian, 21 February 2013.

3 Cheryl Thomas, ‘Are juries fair?’, Ministry of Justice Research Series 1/10, February 2010. []




Occasionally the mutual incomprehension between jurors and lawyers becomes too hard to ignore, but the legal establishment convinces itself that such things are exceptional and blames the jury for ‘fundamental deficits’, not their precious system.

The phrase ‘absolutely fundamental deficits in understanding’ was used by Mr Justice Sweeney in February 2013 when he dismissed a jury in a high-profile criminal case which had caused the downfall of a cabinet minister. Vicky Pryce, the former wife of energy secretary Chris Huhne, had been charged with perverting the course of justice after agreeing to take speeding penalty points for her husband. The jury had posed ten questions to the judge suggesting deep divisions among members about how they should proceed. Some lawyers insisted the case was ‘unprecedented’ but others acknowledged that it signified a general problem. Two senior lawyers, the former DPP Lord Macdonald and the former Lord Chief Justice Lord Woolf, called for more research on juries.




In America, the largest study ever undertaken on juries found that judges hearing the same evidence considered many of the verdicts perverse, including a majority of acquittals.

In the US the University of Chicago carried out the largest studies on juries in the 1950s and ’60s, culminating in The American Jury by Kalvern & Zeisal in 1966. They cited over 3,500 trials, over 1,000 of which resulted in jury acquittals. The judges thought 600 of those acquittals (57 per cent of them) were perverse, and thought the disparity was due to lack of formal jury training. It is be better to err on the side of acquittal so far as defendants are concerned, but that is not necessarily true for victims, nor those who will bear the loss, injuries or bereavements of future crimes due to wrongful acquittals. Nor is acquitting guilty people the best way of holding oppressive or inefficient judicial systems to account. [See also:]




Much is made of the fact that juries go way back in history, far pre-dating Magna Carta, but less often is it conceded that we have neutered and infantilised them.

Jury trial can be traced back at least to ancient Greece but had its modern origins in the Norman conquest of 1066 and the introduction of reforms by William the Conqueror’s successors in the twelfth and thirteenth centuries. King Henry I (1100–35) and King Henry II (1154–89) established travelling courts, and local citizens (based on shire moots, the precursor of what are now called grand juries) were recruited to round up suspects. A defendant could establish his innocence by ‘wager of law’ (sometimes called ‘compurgation’) or ‘trial by ordeal’. In a wager of law he had to find eleven other people (making twelve including the accused, the forerunner of the petit jury) who would attest to his good character. Thus the very essence of a jury was that its members were very well acquainted with the defendant and the circumstances. Trial by ordeal was more usual in serious offences; occasionally it involved trial by battle between the accuser and defendant (or their hired champions), but often used water to test the will of God: if the defendant was accepted by the water he would sink, and if he was rejected because of his sins he would float and thus be guilty. In 1215 the Pope (Innocent III) banned trial by ordeal, and that same year rebellious English nobility forced Magna Carta upon King John, limiting the rights of the monarch: ‘No free man shall be taken or imprisoned, or deseized, or outlawed or banished, or anyways destroyed; nor will we pass upon him, or commit him to prison, unless by the legal judgment of his peers, or by the law of the land.’ This was far less democratic than popular interpretations make it seem: the word peers was used in the Latin sense for equals, meaning the barons would judge themselves and not be judged by others. They had little concern for the rights of ordinary men and women. However, the jury system grew and as tight-knit social communities gave way to urbanisation, slowly juries came to judge the facts rather than act as accusers or character witnesses. Many prisoners thought jury trial was unfair and refused to submit to it, and in order to compel them to comply they were placed between two boards, and weights were applied until they ‘consented’ or were crushed to death. The last recorded fatal crushing was in Cambridge in 1741. Many jurors were also reluctant to serve, fearing retribution from those they found guilty or from victims enraged by an acquittal – and some feared divine vengeance too if they got a verdict wrong. From the 1780s they were permitted ‘reasonable doubt’ for a conviction (but not for an acquittal). Whether beyond reasonable doubt meant beyond all reasonable doubt led to centuries of confusion until 2020, after which English and Welsh jurors were told they must be ‘satisfied so that they are sure’. Whether being satisfied one is sure is different from being sure (or less confusing that ‘beyond reasonable doubt’) was a matter of opinion. As is typical in criminal law, no research was undertaken to establish whether the new wording was any less confusing than the eighteenth century one.




Many people who have served on juries will swear that they were scrupulously fair, while others complain that their experiences appalled them.

In Britain we can only evaluate our juries on through anecdote or rely on scant evidence from so-called ‘shadow juries’ which suggest that at best the system is rather like the curate’s egg, good in parts. Among problems identified were that juries tend to be coloured by pre-publicity and place over-reliance on weak evidence (notably what witnesses say) even where experts consider such evidence to have been discredited. There has been research elsewhere, such as Canada and New Zealand, which suggests juries are generally diligent and fair-minded and are ‘usually collectively willing and able to determine cases on the evidence and in accordance with the law’, but are ‘inefficient’, and ‘some jurors do not engage in a rational consideration of the evidence or become distracted by irrelevancies’. (See Summing-up to Juries in Criminal Cases—What Jury Research says about Current Rules and Practice by a New Zealand High Court judge, William Young, Criminal Law Review (2003) pp. 665–89. This article has a comprehensive list of references. []).




… it is official policy to maintain the secrecy of what goes on and jurors can be sent to prison if they speak about their experiences – even if they feared for the ‘propriety of the verdict’ and acted in good faith.1 The 1981 Contempt of Court Act went further and banned formal research on juries.

1In November 2012 a juror was imprisoned for four months for telling a defendant he thought the verdicts had been reached on perverse grounds. According to the High Court, while he ‘may have acted with a very genuine concern, he took the wrong action’.

2 Section 8 of the Contempt of Court Act was a rushed through after one of Britain’s most famous trials when the leader of the Liberal Party, Jeremy Thorpe, was acquitted of conspiracy to murder a potential blackmailer, and a member of the jury gave an interview explaining how they had reached their decision. In 1993 a royal commission proposed relaxation of the law to facilitate research but in 2005 the government decided no immediate change was necessary.





There is a widely held view that juries are democracy in action and a bulwark of freedom against abuse and tyranny.

One of Britain’s most eminent judges, Lord Devlin, famously described juries as ‘a little Parliament’, and as ‘the lamp that shows that freedom lives’.




… when Lord Carlile QC, a Liberal Democrat politician and human rights lawyer, was appointed by the British government to review terrorist legislation he reported, ‘There is general and justified agreement that the quality of judgement in the Diplock courts is very high.’

Letter from Lord Carlile of Berriew QC, April 2006, Replacement Arrangements for the Diplock Court System, A Consultation Paper, NI Office, August 2006, Annex A, p. 19. According to Lord Carlile, the courts worked so well that they created wide public acceptance and consensus: ‘I have yet to discover any high level of interest in the Diplock courts issue outside the community of politicians, lawyers, academics, interest groups and lobbyists.’ Notwithstanding this, Lord Carlile is a traditional lawyer and defends the jury system in the face of his own evidence.




As a result Crown Court trial without jury was permitted in England and Wales, though so far only in exceptional cases.

Powers came into force in 2007 to allow jury-less trials in England and Wales where there was a real and pressing danger of tampering with juries (Sections 44 and 46 of the Criminal Justice Act 2003) and in 2010, in the first Crown Court case of its kind in 350 years, four men were found guilty of a £1.75 million armed robbery at Heathrow. This was the fourth trial for one of the defendants, the third for one and the second for another. The final trial before a jury had been halted in 2008 when the judge dismissed the Old Bailey jury, saying, ‘It wouldn’t be appropriate for me to give you any further information about what has occurred.’ By then the court costs had exceeded £22 million.




As a means of tackling crime, courts are also long-winded. The government has acknowledged that ‘unforgiveable delays’ are endemic

Damian Green, Policing and Criminal Justice Minister Damian Green, 19 February 2013. []




Defendants held on remand can expect to kick their heels for four months on average; those on bail a good deal more; meanwhile witnesses and victims are in the doldrums too.

Judicial and Court Statistics 2011, Ministry of Justice, 28 June 2012, p. 44. Whether or not justice delayed is justice denied, these long-drawn-out procedures must surely undermine one of the stated purposes of justice, which is to teach people a lesson. Imagine touching a hot stove and not feeling the consequence for between three and six months. []




In any case, for the most part juries are an irrelevance. Some 97 per cent of criminal cases that come before the British courts are heard by magistrates.

Actually the proportion of trials heard before a jury is only around 0.8 per cent because over 70 per cent of defendants eventually plead guilty: Judicial Studies Annual Report 2004, London, The Stationery Office, 2005.




‘Words were put into the mouths of men who had been silent spectators during the whole short episode; actions were attributed to the chief participants of which not the slightest trace existed; and essential parts of the tragi-comedy were completely eliminated from the memory of a number of witnesses.’

M. (Etienne) Dumont, A Treatise on Judicial Evidence, Extracted from the Manuscripts of Jeremy Bentham, London, 1825, and Rationale of Judicial Evidence, ed. John Stuart Mill, (5 vols), 1827.




We don’t see and record things like a camera. Our brains are guessing and inventing all the time, struggling to make sense of impulses from our eyes and ears that represent the world around us.

Consciousness is a virtual reality. As Democritus spotted precociously some 2,500 years ago, ‘Ostensibly there is colour, ostensibly taste, smell; in reality only atoms and the void.’ Our ability to create meaning from it all requires the brain to impose its own interpretation. We can see it struggling to do so in optical illusions but all our perceptions involve construing clarity from ambiguity – and the penalty for our skill at inference is that we sometimes get the wrong end of the stick. Moreover, the brain consumes an enormous amount of metabolic energy (it makes up 2 per cent of body weight but takes up 20 per cent of calories) and so it screens out or compresses what it can. We often fail to notice important things, let alone remember them. One of the most entertaining illustrations of this is the basketball video where if viewers focus on the game they fail to see a man in a gorilla suit coming centre stage and waving at the camera. [].







Even the process of retrieving the memory will alter the way it is stockpiled and later recalled.

This appears to be because of the way our brains store recollections in synapses, the tiny gaps where one nerve branch meets another, and each stimulation of a synapse destabilises it, switching it from ‘read only’ mode to ‘edit mode’. Hence neuroscientists have shown how fallible memory is by manipulating it at the cellular level. By activating brain cells artificially they could make animals transpose recollections so that mice would confuse a safe cage with one where they had been given electric shocks. (Steve Ramirez1, Xu Liu, Pei-Ann Lin, Junghyup Suh, Michele Pignatelli, Roger Redondo, Tomás Ryan and Susumu Tonegawa, ‘Creating a False Memory in the Hippocampus’, Science, 26 July 2013, Vol. 341 no. 6144 pp 387-391 [DOI: 10.1126/science.1239073].)

Manipulating memory like this can be useful. For example, psychiatrists hope to treat post-traumatic stress disorders by administering drugs (stress-relieving beta blockers seem promising) before asking patients to recall the distressing events in counselling sessions. Once recalled with the calming effects of the medicine, the memories may be re-filed as less harrowing ones. On the other hand replanting memories can be very dangerous. Various proponents of ‘regression’, ‘re-birthing’ and other quack methods have claimed to ‘recover’ suppressed memories of ‘child abuse’ and other childhood traumas. More likely the practitioner is helping to paint in memories of events which never happened. See False Alarm: The Truth about the Epidemic of Fear by Marc Siegel, John Wiley & Sons, 2005.




Liszt’s demonstration has been repeated in dozens of forms ever since

Way back in 1908 the pioneering applied psychologist Hugo Munsterberg was able to assemble a range of evidence to challenge the convention that witnesses can be relied on, but perhaps quite as telling as the experiments was his own experience in court: ‘Last summer I had to face a jury as witness in a trial. While I was with my family at the seashore my city house had been burglarised and I was called upon to give an account of my findings against the culprit whom they had caught with a part of the booty. I reported under oath that the burglars had entered through a cellar window, and then described what rooms they had visited. To prove, in answer to a direct question, that they had been there at night, I told that I had found drops of candle wax on the second floor. To show that they intended to return, I reported that they had left a large mantel clock, packed in wrapping paper, on the dining-room table. Finally, as to the amount of clothes which they had taken, I asserted that the burglars did not get more than a specified list which I had given the police. Only a few days later I found that every one of these statements was wrong. They had not entered through the window, but had broken the lock of the cellar door; the clock was not packed by them in wrapping paper, but in a tablecloth; the candle droppings were not on the second floor, but in the attic; the list of lost garments was to be increased by seven more pieces; and while my story under oath spoke always of two burglars, I do not know that there was more than one. How did all those mistakes occur?’ Hugo Munsterberg, On the Witness Stand: Essays on Psychology and Crime, 1908 [available online at].




Professor Loftus later conducted the biggest false memory test ever and showed that half of us can easily be convinced we saw things that never happened.

Steven Frenda, Eric Knowles, William Saletan and Elizabeth Loftus, ‘False Memories of Fabricated Political Events’, Journal of Experimental Social Psychology, 2013, Vol. 49, pp. 280–86.




She summed up three decades of research thus: ‘People’s memories are not only the sum of all that they have done, but … also the sum of what they have thought, what they have been told, what they believe. We seem to reinvent our memories, and in doing so we become the person of our own imagination.’

Elizabeth F. Loftus, Make-believe memories, American Psychologist, Vol. 58, No. 11, 2003, p. 872.

More recently researchers have even persuaded a majority of people to admit to crimes which never happened, including serious assault with a weapon. University students were interviewed about their teenage years and were then asked about incidents supposedly reported by their parents, but actually made up. Most students failed to recall their ‘crimes’ at first, but by the end of their third interview 70% were able to remember them in detail. (Julia Shaw and Stephen Porter, ‘Constructing Rich False Memories of Committing Crime’, Psychological Science, 14 January 2015.)




Six Dutch psychologists explored people’s recall of the most explosive moment in contemporary Dutch history, the assassination of the flamboyant politician Pim Fortuyn in 2002. Almost two-thirds of the people questioned recalled the actual footage of the murder, and almost a quarter were able to remember details from the video. The only problem was: the assassination was not caught on camera. All these ‘memories’ were false.

Marko Jelicic, Tom Smeets, Maarten J. V. Peters, Ingrid Candel, Robert Horselenberg and Harald Merckelbach, Assassination of a Controversial Politician: Remembering Details from Another Non-Existent Film, Appl. Cognit. Psychol. 20: 591–96, 2006. A similar study found that 44 per cent of people recalled seeing non-existent footage of the collision which killed Diana, Princess of Wales. These findings seem broadly consistent. Between 38 per cent and 66 per cent routinely ‘see’ disaster videos that were never filmed, like the jumbo jet that crashed into an Amsterdam apartment block and the sinking of the ferry Estonia far out at sea in which 900 people died.

We are also surprisingly fallible in recognising people’s faces. In 2014 psychologists showed that even trained immigration officers failed to spot that 15% of the people standing in front of them were not the same as the photos in their passports. The error rate rose to 20% if the passport pictures were 2 years old or more. David White, Rcd Kemp, Rob Jenkins, Michael Matheson, Mike Burton, “Passport Officers’ Errors in Face Matching,” PLoS ONE 9(8): e103510, 2014. [doi:10.1371/journal.pone.0103510 or]




London’s former police chief Sir Ian Blair showed how memory can be fallible even for a copper who is experienced with evidence and is on the record. ‘It was a very defining moment,’ he said, describing how as a young constable he came face to face with IRA men in London in 1976 who opened fire on him. ‘It was just a sudden astonishing moment when you actually hear the bullets going over your head.’

Quoted in The Guardian, London, 30 March 2007.




Not that we are any better at judging when people tell the truth. Psychologists have been testing this for years and, as one of them puts it, ‘The results have been remarkably consistent – when it comes to lie detection, the public might as well simply toss a coin.’

Richard Wiseman, Quirkology, Macmillan, London, 2007.




In 2005 psychologists took old fingerprints that had already been matched and gave them back to the same experts. But this time they implied the prints were simply for elimination. Given that context, four of the five experts changed their opinion – reporting that the prints did not match even though five years previously they had said that they did.1 The experiment was subsequently repeated on a much larger scale with similar results.2

1Contextual information renders experts vulnerable to making erroneous identifications, Itiel E. Dror, David Charlton, Ailsa E Pe´ron, Forensic Science International 156 (2006) 74–8 [See also:]

2Itiel E. Dror and David Charlton, Why Experts Make Errors, Journal of Forensic Identification, 600 / 56 (4), 2006. As the researchers point out, fingerprinting is reliable; it is humans that are unreliable. Itiel Dror is a psychologist at University College London who has gone on to be deeply critical of courtroom ‘ignorance’ of cognitive bias in general, but also the vulnerability of scientists themselves to unconscious predispositions. He has given evidence to both UK and US governments. The lesson for fingerprinting and other forensic analysis is that, wherever possible, all samples for testing should be submitted context-free.

In 2009 the US National Academy of Sciences issued a trenchant warning: “The simple reality is that the interpretation of forensic evidence is not always based on scientific studies to determine its validity. This is a serious problem. Although research has been done in some of the disciplines, there is a notable dearth of peer-reviewed, published studies establishing the scientific basis and validity of many forensic methods. (Strengthening Forensic Services in the United States: A Path Forward, Nat Academy of Sciences, 2009, p8




… judges and magistrates do not read the scientific literature and juries have no training at all.

In a famous experiment back in the 1950s, often repeated since, the American psychologist Solomon Asch showed how people who spontaneously gave the right answer to a question would change their minds to fit in with the majority, and a third of them would do so even if their answer was obviously wrong. As ancient China’s wily Xhao Gao observed, ‘point at a deer and call it a horse’ and a lot of people will believe you.

More disturbingly, everyone involved in the judicial process can be innumerate. Two mathematicians have compiled a compendium of notorious cases where lawyers have got the wrong end of the statistical stick and where, as the authors put it, people lives were, ‘ripped apart by simple mathematical errors’ (Leila Schneps and Coralie Colmez, Math on Trial: how numbers get used and abused in the courtroom, Basic Books, London, 2013). They include the wrongful conviction of Sally Clark (who was convicted of killing two of her sons in 1999 on the basis of spurious arithmetic) and the murder of Meredith Kircher in 2007 (where DNA evidence implicating her housemate Amanda Knox was disallowed because the judge failed to grasp the probabilities – see




There was a time when arrest, charge, trial and even hanging followed in quick succession.

In the eighteenth century ordinary trials might last a matter of minutes. There were no lawyers and defendants were not given advance notice of the evidence because it was assumed their spontaneous reactions would indicate their innocence or guilt. However plaintiffs were also disadvantaged, having to bring, and pay for, the prosecutions; and many sentences were never carried out. For a brief history – albeit from a perspective wholly uncritical of contemporary legal processes – see the Old Bailey’s website [].




In 1910 the trial of Henry Hawley Crippen gripped the nation and, as was typical of the time, the trial took four days, the verdict twenty-seven minutes and, even allowing time for an appeal, his execution followed four weeks later.

Remains of Dr Crippen’s wife Cora were found buried in their cellar and, as Henry Crippen and his lover fled to Canada the authorities were tipped off in a world first for ship-to-shore radio. It was a challenging case for the prosecution, not least since the remains were hard to identify, and Crippen denied the charges even on the gallows.




In a gruesome and complex case, Britain’s most notorious post-war killer, John Christie, was arrested at the end of March 1953, was in the dock of the Old Bailey ten weeks later and, after a four-day trial and eighty-two minutes of jury deliberation, he was hanged within a month.

John Christie was later shown to have murdered seven people at 10 Rillington Place in west London, exposing one of Britain’s most celebrated miscarriages of justice. Three years earlier, in what had seemed a straightforward domestic homicide, another tenant there, Timothy Evans, had been executed after the bodies of his wife and baby had been discovered in the house. Evans was posthumously pardoned leading to an outcry against capital punishment. It is tempting to cite such wrongful convictions as justification for longer trials. Indeed, the other notorious case which also helped hasten an end to hanging in Britain was that of Derek Bentley, whose trial for the murder of a police officer lasted less than three days and whose jury took only forty-five minutes to find him guilty. But there is no evidence that miscarriages of justice are less frequent now than then, and arguably Britain’s abolition of the death penalty has made wrongful convictions rather less oppressive.




Nowadays important cases face so much delay that many trials have to be abandoned

In England and Wales the inspectors of the Crown Prosecution Service found almost 3,500 cases had been abandoned in 2005–6 and 2,500 more in 2006–7 because repeated adjournments were deemed unfair to defendants.




It is not even clear that lawyers are always necessary. The only randomised trial ever conducted into their efficiency was in civil not criminal hearings, and in tribunals rather than courts, but returned the ‘unexpected’ result that their involvement caused expensive delays and that their clients would have been better off without them.

This was a trial involving civil not criminal law and before tribunals rather than courts, and since the experiment is unique and its findings so surprising the Harvard authors ‘caution against both overgeneralization and undergeneralization’. Nonetheless they acknowledge that ‘we cannot as yet make reliable generalizations about the circumstances under which representation makes a positive difference’. D. James Greiner & Cassandra Wolos Pattanayak, ‘Randomized Evaluation in Legal Assistance: What Difference Does Representation (Offer and Actual Use) Make?’ The Yale Law Journal, July 2012 p. 2118–2201. [–what-difference-does-representation-%28offer–and-actual-use%29-make?/] or []




In 2003, for example, the double jeopardy rule was finally abolished; for 800 years it had suppressed any new evidence from being heard once a case had been through court.

The 800-year-old double jeopardy rule barred someone being tried for the same offence after an acquittal. No doubt preventing unjustly repetitive arrest was an important safeguard in the thirteenth century, but eventually it was argued that no sane person should refuse to reconsider facts when confronted with new evidence and nor should the courts. The issue came to a head over a violent man called Billy Dunlop. In 1991 he had been cleared of the murder of Julie Hogg, whose body had been found behind bath panelling two years earlier. Dunlop was later imprisoned for attacking a girlfriend and got a further six-year sentence for perjury after boasting of killing Julie. Julie’s mother campaigned to have the double jeopardy law repealed and the Criminal Justice Act of 2003 allowed the Director of Public Prosecutions to authorise a new trial. There was a further significant reform in 2006 when, for the first time in 200 years and to the dismay of old-school purists, some English prosecutors were allowed to interview victims and other witnesses before trials. The practice had been banned in the nineteenth century as a safeguard against witnesses being bribed by private prosecutors, though, as a Canadian lawyer asked in bewilderment, ‘How on earth can prosecutors make informed decisions without speaking to victims and witnesses? ’




This is vividly expressed in verdicts. Unlike an inquest, which seeks the truth, a criminal justice system only pursues a crude conclusion: guilt or innocence.

The test in England, Wales and Northern Ireland, ‘beyond all reasonable doubt’, emerged from a celebrated case Woolmington v. DPP in 1935. Since nothing is beyond all reasonable doubt, that was watered down to ‘beyond all doubt’, or ‘a high degree of probability but not proof beyond a shadow of a doubt’. Nowadays juries are supposed to be told they must be: ‘sure of it. Nothing less than that will do. If after considering all the evidence you are sure that the defendant is guilty, you must return a verdict of guilty. If you are not sure, your verdict must be not guilty.’ Not infrequently they go back to the judge to ask for clarification. No wonder so many juries are confused and, perhaps, so many guilty are acquitted. In Scotland the test is ‘beyond reasonable doubt’ – but for 300 years Scotland has retained a not proven verdict, which effectively means, ‘we know he did it but there is not enough proof to send him to prison’ – or at very least, ‘he might not have done this crime but given its moral connotations we cannot say he is not guilty’. Defenders say it highlights inadequate prosecutors and properly allows a cloud to remain over someone who, in the court’s view, has been lucky to be acquitted. Opponents say it is archaic and unfair, and it was famously derided by Sir Walter Scott as ‘that bastard verdict’. Even so it is used in one-third of acquittals by juries and a fifth of acquittals by lower courts. There has been talk of introducing it in England and Wales for cases where guilty verdicts are later overturned on appeal because of technicalities.




It is a splendid principle that rather ten guilty persons should escape than that one innocent should suffer.

The expression ‘better that ten guilty persons escape than that one innocent suffer’ is generally ascribed to Sir William Blackstone, though the phrase pre-dates him. Blackstone was an eighteenth-century English legal commentator, MP and judge who is accorded absurd respect by some modern lawyers. The philosopher Jeremy Bentham described him as the enemy of all reform. He regarded ‘papists’ as fifth columnists and wrote, in the most gracious prose, that the laws of nature and those of parliament are merely different examples of the same principles ordained by God. For all that, his four books (or Commentaries on the Laws of England) are still quoted reverentially, especially in the US.




Unless they were needed as witnesses, victims had no role in the proceedings.

On the other hand, if they will not appear as witnesses the case may well collapse. This is a frequent problem with sex assaults and domestic violence

Take a typical domestic violence in which the prosecutors finally persuaded the aggrieved to go to court. She had been repeatedly beaten, locked up, bullied and stalked by a boyfriend who had already been to prison for attacking his previous partner. He was arrested a remanded to a bail hostel, but he was highly manipulative, sending her streams of text messages and making her so frightened she didn’t dare change her phone number in case when he got out that would prompt him to seek her out in person. The court agreed that she could give evidence by video link, but even that was more than she could bear: ‘I just couldn’t have him staring at me.’ The case collapsed, he was freed, and she fled home and took her children into hiding in a women’s refuge. Yet there should have been no need for her to be involved in court at all. There was evidence galore about her terrified calls to the police, the scenes found by officers when they arrived, the hospital records of her injuries, the defendant’s long history of violence … but all this was abandoned.




There is no general duty in English law to give evidence against another citizen in a criminal trial, so in general the system relies on people’s sense of goodwill and civic duty.

Under the Police & Criminal Evidence Act, 1984, witnesses can be compelled to give evidence in cases of child abuse or domestic violence, but the provision is rarely invoked. There is an additional set of problems with professional witnesses In November 2004 the Attorney-General, Lord Goldsmith, warned that some paid witnesses were ‘charlatans’, and the chairman of the Criminal Cases Review Commission Professor Graham Zellick, said some paid experts seemed to ‘make it up as they go along’. The selection of professional witnesses is itself not conducive to honesty. I was once offered several thousand pounds for evidence (albeit in a civil case) – but, of course, there was a catch: the job was open only if I said exactly what they wanted me to.




Yet the criminal justice system has traditionally treated witnesses with such high-handed indifference that many are deterred from coming forward.

Sarah Spence and Beatrice Stern (with a foreword by Nick Ross), Reluctant Witnesses, Institute for Public Policy Research, London, 2001.




The charity Victim Support had a huge impact and procedures were overhauled in 2003 after the Law Commission acknowledged that the rules of evidence were defective (often exposing witnesses ‘to gratuitous and humiliating exposure of long-forgotten misconduct’).

Criminal Justice Act 2003, sections 98–113. Among improvements were limits to questions about a witness’s sexual history. But until the reforms in the early 2000s the system was generally so poor that it was difficult to speak openly about its failings for fear of deterring even more people from coming forward. Witnesses and victims were rarely kept informed of progress in the case, implied promises of anonymity were sometimes broken, and unless they remained in court throughout the trial frequently nobody bothered to tell them about the verdict or the sentence. Since then far more information has been offered to victims and witnesses, TV links can be provided for vulnerable witnesses, and courts can sometimes accept written statements where witnesses are too frightened to appear at all. In extreme circumstances a witness can remain anonymous. Targets have been set to reduce victim waiting times, and national standards have been set to improve conditions generally.




It is hard to exaggerate the extent to which the criminal justice process is dissociated from the business of protecting victims and cutting crime. It measures its impact on the world by measuring itself, and even then, as one official report noted, ‘each part of the system has little regard for the consequences of its actions on the other parts’.

Patrick Carter, The Carter Report: Managing Offenders, Reducing Crime, A New Approach, December 2003.