The first time I became really interested in miscarriages of justice was the Birmingham Six. I was aware of the explosions in Birmingham in 1974, and aware of the subsequent convictions, yet repeated stories of problems with the conviction resulted in their convictions being quashed in 1991. I was fascinated. How did that happen? In short, the accusation was that there was nitro-glycerine on their hands. They insisted they had been playing cards. It turned out that playing cards can provide a positive result for nitro-glycerine in presumptive testing and that further testing had not been conducted, either by error or intent.
There was another case in the 1950s where a young petty-criminal, Derek Bentley, shouted to his colleague with a gun “Let them have it”. The court decided Bentley meant “Shoot” rather than “let them have the gun”. Bentley was hanged. The case re-emerged in the context of the ending of the death penalty in the 1960s.
More recently, many more cases of miscarriage of Justice have come to light: the NHS blood scandal, the Post Office scandal, supposedly the biggest miscarriage of justice in English legal history, Hillsborough (although that case may just be prejudice) as well as others.
But, it was in the USA that a sudden surge of acquittals emerged as a result of DNA tests (Cutler, p14). This was well-covered on TV documentaries including the Innocence Project on Netflix, but, notwithstanding Nickerson (1998) who laid the groundwork for the recognition of the “ubiquity” of confirmation bias, Findley and Scott (2006) was truly seminal in the social as well as the forensic sense: that so many people can believe someone is guilty when they aren’t and refuse to accept they were incorrect from the bottom of the legal hierarchy to its pinnacle. The Birmingham 6 were convicted because everyone “believed” they were guilty when they weren’t. This “belief” starts as individual confirmation bias and morphs into corporate groupthink. It seems to be compounded by an inherent belief by civil servants or other “corporate” style workers that what a fellow civil servant wrote must be true. In the social setting confirmation bias plays out subconsciously with all the players or agents everywhere. Miscarriage of justice is a legal setting. That is what we are considering here.
What is confirmation bias?
“While anybody is free to approach a scientific inquiry in any fashion they choose, they cannot properly describe the methodology as scientific, if they start with the conclusion and refuse to change it regardless of the evidence developed during the course of the investigation.”— William Overton, McLean v. Arkansas 1982, at the end of section IV. (C)
Wason (1960) came up with the name confirmation bias, but Nickerson (ibid; p176) originates bias in England with Francis Bacon:
“The human understanding, when it has once adopted an opinion draws all things … to support and agree with it. And though there be a greater number and weight of instances to be found on the other side, yet these it either neglects and despises or … sets aside and rejects in order that by this great and pernicious predetermination, the authority of its former conclusions may remain inviolate”.
Bacon was the patron of my school as it happens, but there were Greeks who understood the concept too, as well as Romans. The key factor is that bias is unconscious. Prejudice is conscious. I prefer to use the word sub-conscious because I consider these biases are closer to the surface than we think.
Wason suggests that confirmation bias occurs when people seek out or evaluate information in a way that fits in with their existing thinking and preconceptions. That’s a modern way of putting it, but in evolutionary terms it seems obvious that choosing things that give you pleasure is rational that is, logical, in its own way.
Then, we ask, does a prosecutor get pleasure in prosecuting in itself, prosecuting successfully or winning the case when they even knew it was unsound? Is that rational in legal terms? I would think not. That is self-serving. Hence, the logic of the behaviour is context-driven. Academics might call that social constructionism and as a social psychologist I would say environmental influences play a constant part. Or, another example, in social psychology we have in-group and out-group theory, not least minimal groups that show that just putting individuals in the same group creates a bias towards that person. So it goes on. These processes are sub-conscious: we don’t know we’re doing it.
The Process 1
So, in overtly Socratic form, what are the effects of confirmation bias on the courtroom setting, starting with the process up to the court setting and why do the CPS have a 50% failure rate in prosecutions?
Rather than Britain, it seems we must look to the USA for social evidence and Butler’s “Conviction of the Innocent” (2013). It is an incredible read, albeit rather expensive to buy. From the current perspective two figures stand out: Findley’s “Tunnel vision” and Itiel Dror’s work on bias in the forensic sciences. The Americans use the phrase tunnel vision to mean confirmation bias, quite understandably, but I will continue with confirmation bias.
Taking up the social thread, from there we can go to Findlay and Scott’s work (2006), and further, Nickerson (1998) the ubiquity of confirmation bias, which is at times a hilarious romp through examples of confirmation bias through history (e.g. p194).
Historically, bias was certainly understood by the Greeks, but in England Francis Bacon made it clear bias was understood. Man imposes his own view on his perception of the natural world. From there a new mind-set emerged: objectivity. One must identify the sub-conscious self-interest in one’s own thinking. More specifically, there was a historical process of thought:
1290/91 and Edward I’s “Quia Emptores”, which allowed private ownership of land for the first time in history,
the Renaissance c1300, which gave us perspective in art and later the ”banking” system,
the Battle of Bosworth Field 1485 which gave us the accountant King Henry VII,
the Reformation and the rise of the individualistic Protestant work ethic,
the Civil war, which ended absolute monarchy, and, finally,
the Enlightenment where, courtesy of Benjamin Franklin in 1857, with his kite and key, God was removed from the equation and statistical probability replaced Him.
Apparently, Huygens wrote the first book on probability in 1657 in the context of gambling: “De Ratiociniis in Ludo Aleae (On Reasoning in Games and chance)”.
In this way, the English and Scots, led by Hume, broke the chain of confirmation bias and created the objective and, dare I say, scientific mind. A new mindset. It is mindset that determines the personality of an individual and mindset is particularly subject to socialisation, culture and, in the West at least, “positive testing”. Tunnel vision is to some extent schooled by the conventional education system through “positive testing” (Klayman and Ha 1987; Narveson 1980; Nickerson 1985) despite the fact that science should be sceptical and questioning and aim to disprove facts or accepted ideas. The peasant world that preceded the Enlightenment and most certainly still survives as the “global south” is very different socially from the “disenchantment” if not of the world, as Weber suggested, but the West; the white, Anglo-Saxon, Protestant West which created the capitalist-individualistic world we see now (See Alan McFarlane; The Culture of Capitalism (1987).
By this time, 1620, Socratic questioning reappeared after 1,000 years of university-led Aristotelianism, the bureaucrat, the follower of process, the bureaucratic mindset. The possibility of a questioning, Socratic mind established itself and in turn created innovation. Capitalism prospered even if the Law did not proceed so well.
The 1688 Glorious revolution saw free speech enshrined. Through Kant we started to understand perception through ““Geist” und “Schein””. By 1800 the English were a nation of shopkeepers and as any bourgeois capitalist shopkeeper can tell you, if you’re biased you won’t last long in business. But confirmation bias still persisted. It persisted in institutions in particular.
It wasn’t until DNA testing appeared in the mid-1980s that the dire consequences of a failure to properly investigate crime was exposed and threw light on peoples’ false perceptions guided by their own, subconscious, thinking and preconceptions.
It is here we must give full credit to American researchers.
The Process 2
All that being the case, we can create a scenario. A man, Mr. X is arrested. He is charged. The papers are assembled. He may be bailed. He may not. He is interrogated. He has a defence lawyer, a solicitor, with him. His defence lawyer suggests he pleads guilty. He doesn’t. It eventually goes to court before a jury. What happens then? At this point I will cite Findley and Scott at length:
“Defence lawyers are susceptible to institutional pressures that contribute to tunnel vision. Defence lawyers, like prosecutors, quickly learn that most people charged with crimes are guilty. They see that most of their clients plead guilty, and most of those who go to trial are convicted. Defence lawyers learn that, more often than not, their clients fare much better in the criminal justice system if they plea bargain rather than go to trial. To take an adversarial posture in a case, to investigate aggressively, file discovery motions for access to evidence from police and prosecutor’s files, and file motions to supress – not only requires an investment of scarce resources, but also comes at a cost in terms of the ultimate resolution of the case. Experienced defense lawyers learn that better deals can be obtained by being cooperative. “Prosecutors and judges alike thus indoctrinate defense attorneys into the plea bargain process by communicating to attorneys that time-consuming motions should be forsaken in favour of plea negotiation”” (Findley and Scott 2006, p331).
In the UK, of the some 500,000 cases per year in the UK, 67% of people plead guilty (guilty or not), 33% plead not guilty. Of that 33% the CPS prosecute, the CPS only obtain a 50% success rate, as admitted by Max Hill DPP, in Parliamentary committee in 2022. Most statistical examinations would require less than 5% fail rate. Confirmation bias and subsequent groupthink is alive and well in the CPS.
Mr X has already been through layers of possible confirmation bias. Most lawyers consider a person guilty when charged. He has been interrogated aggressively. The police don’t believe him. The defence lawyer does not interject in his defence (Naughton, 2013, p80). The defence lawyer and the detective are old friends. In fact, the defence lawyer used to work for the CPS previously (this is all taken from personal experience). Perhaps he knows the prosecuting barrister. They exchange views on Mr X later. A date is secured to go to court. There is a pre-case hearing. They go to trial.
The Trigger list
Here it becomes simpler: imagine Mr. X, in a legal setting, here in England. Imagine the solicitor and the defence barrister, the latter as instructed by the prosecution, famously the CPS or “Crown Prosecution Service”. Imagine the prosecutor’s mindset. He wants a win. The clerks’ disinterested mindset, the jury’s expectant mindset. The entire body of people are engaged in the focus of attention on one individual (in this case) in the dock, but also picking up social signals in appearance and body language, including facial expressions.
Having already fallen foul of the confirmation bias shown by the lawyers, Mr X now faces the court and jury as a whole: how does the court react to Mr X? At this stage, as in the process leading up to this moment, confirmation bias is, as Nickerson would put it, ubiquitous. It is absolutely everywhere. In effect, people are making un- or sub-conscious judgements about everything and everyone around them. At this point we come to the focus of our attention as to what triggers confirmation bias in a legal setting. On that basis, I reviewed all the evidence I could at that time, and I found the following list. Any of these triggers could act upon an individual that would set-off confirmation bias and unwittingly affect their choice of evidence, or ignoring thereof, that would determine their judgement or decision-making process, either negatively or positively.
In the first instance we can talk about the Primacy Effect (Asch;1946). This means what it says in the old saying of “first impressions”. That you can, in reality, judge a book by its cover. People were pre-warned someone was “warm“ or “cold” and the audience responds accordingly. This initial impression leads to belief perseverance, caused in turn by anchoring: once you’ve made up your mind, you don’t want to change it.
Then we have prior expectations. We think we know what it’s about. Mr X didn’t look quite right to Ms Y in the jury. When she hears the charges, that confirms her first impression. She then only listens to the negative information or what feeds her preconceptions. That is, sui generis, confirmation bias, triggered by the primacy effect
Then we have Social Identity Theory in the form of in-groups and out-groups notably Minimal Groups (Tajfel 1970/79). In this case, social markers are triggers for association or rejection, not least the swirling robes and white wig of barristers. Ethnicity may play a part or sex or gender more recently in the case of Trans people. The actual identity may take time to appear in the court through what Mr X, the defendant says, perhaps from his accent.
Then, in the course of a trial, 3 particular syndromes are identified by Rassin (2010): previous investigations (“no smoke without fire”), crime severity and the quantity of evidence. It has to be borne in mind that evidence is not proof. For example, I have persuasive evidence that the Sun goes round the Earth, and it is there for everyone to observe. The problem is, thanks to the Enlightenment, we know the Sun doesn’t go round the Earth. The crime severity speaks for itself, the more horrible the crime the more vulnerable observers are to falsely attach the crime to Mr X. The quantity and quality of the evidence, “where there’s smoke there’s fire” attitude, which I completely condemn, is itself suspicious. As a professor once said to me: “sometimes the accusation is enough”. I myself have seen fake smoke and it’s called dry ice. As I learnt with my grandfather in my youth, when something happens to a machine in the factory, it could be an accident, a technical failure - or sabotage. In miscarriages of justice, it could be human error as in slips and mistakes, a technical failure would be confirmation bias and sabotage would be prejudice (a stitch-up or “abortion of Justice” (Naughton, 2013)). When the CPS has a 50% failure rate, it’s hard to see that just as a mistake. In fact, taking innocent people to court and acquitting them could be seen as a miscarriage. The punishment is the process.
Dion (1972) - possibly the best example - is where juries and judges have a favourable attitude to attractive people and such people are seen to get lower sentences. Equally they may be more likely to be “believed” even when they are lying. I, myself, have seen a local government agent lying to the court but being believed by the judge, the prosecution and defence lawyer. The fact she was attractive and turned to tears may only have been part of it, when you consider institutional pressures and in-group favouritism. And, all the time it is subconscious.
Institutional pressures are well demonstrated in Findley and Scott’s citation above. Lawyers and judges have to conform to their group opinions and the expectations of the institution or corporation. This confirmation bias in the individual becomes group-think in a corporate setting. Hofstede pointed this yes-man syndrome out in 1980/81 when researching problems at IBM. They all want to agree with the Boss. One has to consider in-group favouritism in the selection of, in this case, law students.
Findley (2012) adds the adversarial system to the mix in that it becomes important not for justice to be seen but victory on the part of the prosecutor or defence lawyer.
Context effects are the same as social construction where environmental factors affect the perception of viewers. There is fundamental attribution error and situation vs disposition where observers over-emphasise the personality of the observed rather than allowing for the situation that is affecting the observed. So, Mr X may appear nervous or agitated because of the courtroom setting when he is, in fact, a mild-mannered and agreeable person. The environment has an effect on his manner. His nerves may look like guilt. On the other hand, Mr X may be a more rough, tough type of character and is unimpressed by the court, perhaps from his own previous experiences or experiences of those he knows which makes him appear in a negative light to some.
Finally, cognitive dissonance (Festinger, 1972) may play its part where the observer came to one decision about Mr X, he appeared a good guy, then when the weight of evidence seems to show another side of the defendant the observer is conflicted. That conflict is hard to deal with and might count against Mr. X.
That is a brief outline of the psychological atmosphere that can surround a defendant in the courtroom, and before, that can sway the decision making of observers.
Dror (2020), lists: data, reference materials, contextual information, base rate, organisational factors, education and training, personal factors and human and cognitive factors, and the human brain in the case of forensic science which has some similarities to the list but is focussed on the forensic science rather than social groups and their thinking.
Disclosure, discovery and abuse of process
It’s worth mentioning that the failure of disclosure by the CPS or police and the difficulty in discovery experienced by defendants can effect the course of the case. In a parliamentary select committee Max Hill, then DPP, while admitting a 50% failure rate in CPS prosecutions, tried to blame it on the police and “abuse of process” which, in fact, means not providing the correct paperwork. One would imagine the CPS had the relevant paperwork before they charged the defendant, unless they simply didn’t want to provide it. It could be that the “process is the punishment” in some cases.
Juries and witnesses
The facts tell us that if 50% of CPS cases fail, then the jury system is working. However, 95% of all cases (some 500,000 per year in the UK) are Magistrates’ courts where there is no jury. 25,000 cases are therefore held in the Crown Court and elsewhere. They are, however, the most serious cases so the role of the jury is important. Disaggregating the numbers, there are about 5 jury cases per week involving people from Dorset. I haven’t researched the court system as such, from a systems level, but it can’t be difficult to work out a schedule. Then we have the jury, the court-room, the barristers (we hope, there is often a shortage of barristers) the witnesses and, we hope, the paperwork.
The facts tell us that juries can be fooled. The facts tell us that witnesses lie.
Conclusion
The conclusion is that lawyers are not to be trusted, certainly not criminal defence or the CPS or, indeed, judges. Other forms of lawyer are not my concern here. Biased thinking is a well-understood idea, but confirmation bias is a specific idea. We have to watch our own behaviour and ask ourselves: is that right? Is it correct? Is it factual? Is it moral?
It may well be that lawyers are trained, in effect through positive testing as mentioned above that you have to prove a hypothesis, not dis-prove it. I went to the debating society, learning how to argue both sides, a lawyer’s training, if you like. The USA, however, as well as my own personal experiences, show the depth of confirmation bias that exists in the judicial system as in any large corporate body. The Post Office. The blood scandal in the NHS. The Grenfell Fire in several governments, it is a damning list.
The university system weeds out unconventional or Socratic thinkers. Socratics ask questions that the interviewee has no answer to or doesn’t want to reply to. Socratic thinking is some 2,400 years old. It isn’t new. The first task of a thinker is to question: is this true? Who provided the evidence? Who is the alleged perpetrator? This is not how the English – or American – system works at this time. The ability of lawyers to collude in order to manage the case, persuading people to plead guilty for example as we saw in the Southport Riots and in the Post Office Scandal – is not what innocent people are looking for. They want their lawyer to fight the case. Findley and Scott explained the deep-rooted confirmation bias that does, indeed, warrant the name “tunnel vision”: hear no evil, see no evil.
We can see there are two, at least, types of miscarriage: a deliberate stitch-up or abortion of justice (Naughton, 2013) and a flawed method or miscarriage of justice. Abortions of justice are deliberate and thus are prejudicial. Miscarriages can be caused purely from the human condition. I would suggest a third type of miscarriage is the innocent proven innocent miscarriage. Humans have confirmation bias naturally, it is innate: it has to be trained out of the mindset. That remains a big ask but one would have thought lawyers would have understood it.
Post-script: a personal account
I grew up in a business family, a capitalist family, we had a factory and shops and still do have an on-line presence. All the men in my family went to technical college and I was sent to management school. Over time it became obvious to me that the main issue in management is decision-making. Psychology is the department that decision-making is studied.
As a mature student, I was 40, I went to Southampton University in 1994 because that was the nearest and my mother was 80 years old by that time and she was dependant on me.
Prior to that time, I had had no interaction with the courts or the police other than the occasional police stop over the years.
While at the department, after the second year, the Head of department, Professor Robert “Bob” Remington accused me of sexually harassing 3 female students. I was told by one of my tutors Remington referred to me as the “posh git”.
I sued the university and won compensation. I knew one of the students from her being on one of the courses, the French student had socialised with me and the other I didn’t even know. That was 1996.
Then in 2000, my house was entered in the morning by police with dogs and I had my house searched. It turned out that someone had (allegedly) left drugs under my garden shed. I didn’t see them. I was cautioned even though it was nothing to do with me. However, the caution gave me a police record. I was poorly advised by my solicitor. I was, some time later, stopped in my car, manhandled, arrested and charged with affray. I was fined £250 and given three points on my licence.
Then in 2005, someone sent me some cash, about £6,000. My bank account was stopped. I was charged with 14 counts of money-laundering. It went to the Crown Court. The judge stopped the case and ordered the jury to acquit. A ridiculous waste of public money.
Then at Christmas 2006, my mother and her husband who both had dementia had a fit or angry episode, simultaneously: Jack fell down and my mother phoned the police. I was arrested. Months later the case went to court. The judge stopped proceedings saying my mother was “not a reliable witness”. I could have told anyone that. Another monumental waste of public money. The case took 2 years to complete with acquittal. That was 2010. It turns out that Boris Johnson’s friend Petronella Wyatt went onto GB News, the UK news channel, and gave an account the same as mine. It wasn’t just me as I discovered through my later research.
Things went quiet. I put it down to PC Plod syndrome, not even really knowing much about the CPS. I noticed that it had seemed to start when I stood for Conservative councillor in Purewell in Christchurch but I moved on.
Then, in October 2018, the police came to my door, searched the house and arrested me. They didn’t say why. I was taken to Weymouth. I was told I was arrested for stalking a waitress in a restaurant in Christchurch (you can Google it). They took my computer and phone. I hired a solicitor. He was ex-CPS it turned out. He didn’t know the brief in court. I was convicted in the magistrates’ court. I sacked him and got compensation. I appealed to the Crown Court representing myself. The CPS withdrew the case: no evidence. Another colossal waste of public money.
Then ,the psychiatrists sectioned me and detained me at the local mental health hospital for 28 days in October 2019. By this time, I knew something was not right. I was released.
However, by this time I was acting as my mother’s carer, with the help of nurses twice a day. One nurse came from the agency Pramacare and she struck me as being somewhat different from the other nurses. It turned out she was a psychology student. She tried to engage me in conversation, particularly about psychology. Something wasn’t right about her. She appeared flirtatious. There was a strange incident with the office of Pramacare which struck me as odd. I sent a Facebook message to the psychologist nurse asking her if she could give any explanation. She had left my “employ” as was her association with me as I had power of attorney. Then I was arrested yet again.
This case dragged on for another year. I was convicted of harassment because I wrote a Facebook message (which seemed the biggest offence) to the nurse asking if she could explain some strange events that took place during the week of April 2020. She didn’t reply. Papers released later (lack of disclosure) show the Pramacare Agency was conspiring with social services regarding the previous case: they were trying to trap me into making advances on the nurse. The fact I wasn’t interested gave the game away.
The upshot was that this female had been co-opted by the local authority and they had asked her to find out what I was like. The council were using Pramacare themselves. I tried to appeal the conviction, but I wasn’t allowed to and then I went to the CCRC (Criminal Cases Review Commission) who told me if I made any more enquiries I would be sued for vexatious litigation. Frankly at this point I’d had enough and paid the £250. I had too much income for legal aid and the cost of proceeding with the case was extreme. The CPS spent all that tax-payers’ money to gain £250.
After complaining to the Parliamentary and Health Ombudsman about my unwarranted detention, it turned out that the Head of the psychology department at Southampton University had had me flagged on the Police National Computer (PNC), one assumes out of spite. This meant I was taken notice of by the local authority. I think it obvious that if in two cases the judge threw both the cases out, the CPS must have problems with their decision-making: it is subject to confirmation bias or worse, outright prejudice.
During the latter cases after 2018 I sent multiple letters to the CPS and the courts pointing out their errors but there was no coherent response.
During the process after 2018 I researched the subject and discovered Findlay and Scott, Itiel Dror and Brian Cutler. Although having studied a lot of sociology and law, my psychology training gave me insight into why apparently rational people could make up evidence and then believe it without causing them cognitive dissonance. The answer is confirmation bias.
Perhaps that is another area for research.
By Robert Luther-Smith (Dip. Man. BA (Hons) (Psych)
Bibliography
Asch, S. E. (1946) 'Forming impressions of personality.' The Journal of Abnormal and Social Psychology. 41(3), 258–290.
Dion, K., Berscheid, E., & Walster, E. (1972) 'What is beautiful is good.' Journal of Personality and Social Psychology. 24(3): 285–290.
Dror. I. (2020) Cognitive and Human Factors in Expert Decision-Making: Six Fallacies and the Eight Sources of Bias. Analytical Chemistry. 2020. 92: 7998-8004.
Festinger, L. (1957) A Theory of Cognitive Dissonance. Stanford, CA: Stanford University Press.
Findley, K.A. and Scott, M.S. (2006) “The multiple dimensions of confirmation bias in criminal investigations”. University of Wisconsin Law School.
Findley, K.A. (2012) 'Tunnel Vision' in Conviction of the Innocent: Lessons from Psychological Research, B. Cutler, ed., APA Press.
Klayman, J. and Ha, Young-Won. (1987) Confirmation, Disconfirmation, and Information in Hypothesis Testing; Center for Decision Research, Graduate School of Business, University of Chicago Narverson (1980); Development and Learning: Complementary or conflicting aims in humanities education; In Fuller et al (Eds) Piagetian programs in higher education. Lincoln, NE: ADAPT program.
Naughton. M. (2013) The Innocent and the Criminal Justice System. A sociological analysis of Miscarriages of Justice. Palgrave Macmillan.
Nickerson, R.S and Perkins, D.N. and Smith, E.E. (1985) The Teaching of Thinking. Psychology Press.
Nickerson, R.S. (1998) Confirmation Bias: A Ubiquitous Phenomenon in Many Guises; Review of General Psychology.
Rassin et al (2010) '“Let’s find the evidence”; an analogue study of confirmation bias in criminal investigations.' Journal of Investigative Psychology and Offender Profiling.
Tajfel, H. (1970) 'Experiments in intergroup discrimination.' Scientific American. 223: 96-102.
Tajfel, H., Turner, J. C., Austin, W. G., & Worchel, S. (1979) 'An integrative theory of intergroup conflict.' Organizational identity: A reader. 56-65.
Comments