Police practice and false confessions: A search for the implementation of investigative interviewing in Australia

The false ‘confession’ of Andrew Mallard

One well-known Australian example of the detrimental consequences of inappropriate questioning techniques is demonstrated in the case against Andrew Mallard. ⁸ Mallard was wrongly convicted for the murder of Pamela Lawrence in Perth’s Mosman Park. ⁹ Lawrence’s body was found in her shop premises in May 1994. Shortly after the WA police started to investigate the murder, they discovered that Mallard was in the surrounding area at the time of the murder. Mallard was charged with burglary in the Mosman Park area and also charged for impersonating a police officer. These offences had taken place the day before, and the day after, the murder. ¹⁰ For the second offence, he was taken to a hospital for a mental assessment. Shortly after his hospital admission, he was brought in for questioning. Mallard initially denied any involvement in the murder. However, in his police interview which lasted over eight hours and of which only 20 minutes were audio recorded, he reportedly made incriminating admissions. There were several concerning circumstances and tactics used during the questioning of Mallard.

First, Mallard was taken for questioning into a ‘formal environment’ (being police headquarters at Curtin House) which had an intimidating effect on Mallard. ¹¹ Subsequent persistent questioning about the gap in his story – a short period of time for which he could not account – affected Mallard mentally. The questioning police officers further highlighted that they were trying to help Mallard and lied about witnesses seeing a suspect matching his appearance, outside the store where the victim was murdered, on the evening of the murder. As a result, Mallard started making declarations about the night of the murder in the third person on what hypothetically could have happened. He later indicated: ‘I just made that up, Sergeant, to get you off my back’. ¹²

Hours after this interview commenced, Mallard was finally cautioned for the first time. By the end of the interview, Mallard became hysterical and violent and was brought back to a hospital to be examined. He was subsequently charged with assault on the interviewing Sergeant. ¹³ At this stage, there was no physical evidence linking Mallard to the crime. However, roughly a week later, the questioning continued. During this interrogation, he reportedly said: ‘Look, I killed her okay, bashed her skull in, took the money and took the purse’. Mallard banged his head and reportedly, subsequently, became a ‘little aggressive’. ¹⁴

Mallard was tried and later convicted for the murder despite the retraction of his alleged admissions and its inconsistencies. ¹⁵ The reported (not audio recorded) admissions constituted ‘confession statements’ and were the main piece of evidence. Several appeals later, it was discovered that prosecution withheld evidence that would have invalidated Mallards statements in the police interviews. ¹⁶ The police also failed to follow up on crucial leads provided by witnesses. The new suspect, Simon Rochford, who was already serving a life sentence for murder, committed suicide in his cell when he overheard that he was the new person of interest in the Lawrence murder case. ¹⁷

The problem of false confessions and interview techniques

Although there were many contributing factors towards Mallard’s wrongful conviction, the key piece of evidence – his ‘confession’ – was false. His wrongful conviction led to a thorough investigation into misconduct in the WA police. The investigation exposed questionable interview techniques relating to the interview with Mallard, who was mentally unstable. In 2009, the then Director of Public Prosecutions (DPP) Robert Cock stood down from his position and stated that the next person in the position should ensure that ‘staff is properly trained’. ¹⁸ Specific elements of the reported interviews with Mallard do indicate an inappropriate line of interrogation. The accusatory line of questioning (in this case, accusing Mallard; fabricating witness statements; and pressuring him) is part of a commonly criticised questioning method previously used by many police forces.

The predominant interrogation method internationally is the Reid method. Many countries, including Australia, have previously used this interrogation method (and variations on it). ¹⁹ The Reid model focusses on interrogation and obtaining a confession, rather than investigative interviewing, and thus information gathering. ²⁰ The coercive and confrontational police questioning techniques used in the model could contribute to a suspect confessing to a crime that they did not commit. ²¹

Mallard’s case study shows the detrimental effects of using inappropriate questioning techniques. With variations of the Reid method used in Australia, the issue of police conduct in interviews with suspects has been a focus of reform. ²² That is, increasing awareness about problems related to this interrogation model, in addition to developments in interviewing techniques in the UK, led to the introduction of the so-called PEACE model in Australia in the early 1990s. ²³ This article will provide an overview of the contributing factors towards false confessions, and analyse and describe which interview techniques are used by Australian police forces. The authors will highlight why investigative interviewing methods that minimise the risk of false confessions are preferable to the questioning methods used in Australia.

The psychology of false confessions

There are various reasons why people falsely confess to a crime. ²⁴ To better understand false confessions, it is important to not only differentiate between types of false confessions but also to acknowledge the different motivations behind them. A false confession occurs when a person admits to committing a crime, or exaggerates their involvement in a crime, while they are actually innocent. Within legal psychological research there is a clear distinction between three types of false confessions, triggered by three different psychological constructs.

A voluntary false confession occurs voluntarily, without being prompted by the police. ²⁵ There are various motivations behind voluntary false confessions. People can falsely confess because they want to protect the real offender (eg, a mother admitting to a crime to protect her child from prosecution). There could be psychological disorders at play – when people confuse reality with fantasy, and thus admit to a fantasy and not to facts. A desire for notoriety or fame could also be a reason to falsely confess to a crime, especially if this crime is widely reported. A famous example of mass voluntary false confessions is the kidnap and murder case of Charles A Lindbergh Jr. In March 1932, 20-month-old Charles Jr was kidnapped from his home in East Amwell, New Jersey, US. Charles was the son of the first pilot to complete a solo transatlantic flight, and reports of his abduction and murder occupied national news for months. It took police over two years to find the alleged perpetrator (Bruno Richard Hauptmann maintained his innocence until he died), while over 200 people reported to the police, confessing to the crime. ²⁶ What this demonstrates is that people (indeed, many of them) can come to confess to a crime they did not commit, without prompting, pressure or coercion. It is important to note that, in the majority of voluntary false confessions, the confessor knows they are innocent, but freely admit to committing the crime.

This is different for the coerced-compliant false confession, which happens when a person falsely confesses to a crime to escape the interrogation, gain a reward, or avoid a threatened punishment. ²⁷ While the confessor realises they are actually innocent (ie, they did not commit the crime), they are coerced into confessing to the crime, due to the circumstances – especially interrogation techniques used. ²⁸ Importantly, by confessing they are complying with the interrogators, while knowing they are innocent. The important psychological construct at play is compliance; conforming to an authority, in this case the interrogator. ²⁹ The confessor reports what they think the interrogator wants them to say, knowing this to be untrue. They want to escape the situation they are in and believe the only way to do this is by confessing. Button’s case, discussed at the start of this article, is a classic example of coerced-compliant false confession.

With a coerced-compliant confession, the confessor is merely complying with the interviewer; however, this changes with a coerced-internalised false confession. Here a confessor comes to believe they committed the crime (while they did not). These types of confessions are often the result of highly suggestive interrogations. They usually start off as compliant false confessions; the confessor is at first aware they did not commit the crime, but then – due to ongoing suggestions – they come to believe they actually did it, by internalising the crime. The psychological construct at play here is suggestibility; the degree to which a memory is vulnerable to distortion. ³⁰ Some people find it difficult to resist suggestions in questions. If suggestions are provided within questions during an interrogation, people can incorporate this information into their memory for an event. Moreover, suggestibility increases when questions are asked by an authority figure. ³¹ Some people, such as those with cognitive disabilities or certain psychological disorders (vulnerable witnesses), are more open to suggestions and thus make coerced-internalised false confessions. While no coercion is needed for a voluntary false confession, the role of interviewing techniques becomes apparent for the latter two types of false confessions. Coerced-compliant, and coerced-internalised false confessions primarily occur due to suggestion and coercive techniques used during police interrogations. The Reid technique is a prime example of such coercive techniques. ³²

Reforming police interviewing techniques

As touched on above, ample case studies have shown the prevalence of false confessions. Since the mid-1980s, recognition of the occurrence of this type of confession led to reform in interviewing practices. Police forces realised they should move the focus away from trying to obtain a confession towards information gathering. As such, practice moved from interrogation towards interviewing.

The UK became a frontrunner in improving interviewing techniques. In 1992, a new interviewing model was developed, informed by academic research on the link between coercive interviewing techniques and false confessions. ³³ The PEACE model, an acronym for the various stages/structure of the interview, ³⁴ is a method known as conversation management, which focuses on conducting an effective and ethical interview. As such, this method specifically encourages information gathering over securing a confession. The model requires interviewers to be well-prepared for the interview they are about to conduct, and to plan out their approach. Interviewers should explain the ground rules and engage the interviewee in the interview. Their final goal is to obtain the interviewee’s account and, if needed, to ask appropriate questions/challenge them. After this, the interview should be properly closed with any remaining issues solved. Last, the interviewer should evaluate the interview, to establish if all goals have been met and whether the interview fits within the bigger investigation. ³⁵ Research assessing the effectiveness of the PEACE model has not only shown a reduction in false confessions but has also shown an improvement for interviewers and interviewees in police interviews as its adoption, for instance, addresses the skills of interviewers and improves general interviewing performance. ³⁶

Moreover, the benefits of the model appear to be ongoing. In a national evaluation survey, Clarke and Milne assessed the PEACE-interviewing skills in the UK. The survey highlighted that approximately 70 per cent of police officers in England and Wales had been trained in the use of the PEACE model. ³⁷ Clarke and Milne concluded that there was a noticeable overall improvement in interviewing behaviour compared to previous criticism on questioning and the provision of legal standards. ³⁸ Although this national evaluation dates back to 2001, the PEACE model is currently the primary questioning method for police interviews in England and Wales and the overarching model for training of police forces. ³⁹ As such the PEACE model has been considered to be good practice by researchers and practitioners, with psychological research considered vital in developing this effective and appropriate way of police interviews with suspects. ⁴⁰

Many countries have chosen to consider the UK developments as a good example to improve the police questioning of suspects in their own country. Some countries chose to directly incorporate the UK PEACE model. New Zealand (NZ), for instance, directly incorporated the PEACE model as a uniform protocol to interview suspects. This was done after a comprehensive review and analysis, which strongly rejected the US interrogation methods and considered investigative interviewing (eg, PEACE model) as good practice suitable for NZ police forces. ⁴¹ Roughly five years after Schollum’s recommendations, the NZ police conducted a national evaluation of the training of police officers in accordance with the PEACE model. It was concluded that the implementation of the PEACE model and accompanying training of the NZ police officers was a notable improvement to police interviews. It also announced that it would continue to actively seek ongoing improvement of investigative interviewing. ⁴² This evaluation also reported that roughly one-third of its staff was trained in accordance with and accredited in the PEACE model. ⁴³ The NZ police is notably transparent in the use of the PEACE model and the training of its police officers. Additionally, the implementation of the PEACE model worldwide has been welcomed and encouraged by the United Nations General Assembly Special Rapporteur when in 2016 he suggested a worldwide protocol for investigative interviewing. ⁴⁴

The extent of reforms in Australia

Like many other countries, Australian police forces used to utilise (elements of) the Reid technique when interviewing suspects. ⁴⁵ Australia has reportedly followed the UK approach ⁴⁶ in investigative interviewing. As a result, most Australian police forces introduced the PEACE model. ⁴⁷ Notably, although the implementation of the PEACE model in the UK is a good step forward, the performance of police officers with regards to specific interview skills can still be improved. ⁴⁸ Queensland introduced its Police Service to the investigative interview principles in the early 1990s; in most other jurisdictions, introduction followed later. ⁴⁹ Although the literature suggests that investigative interviewing has been endorsed nationwide, ⁵⁰ to date Australia has not had a uniform implementation of investigative police interview methods. Hence, the implementation of the PEACE model or investigative interviewing differs for each jurisdiction. While there are indications of a shift to investigative interviewing in Australia, Ord et al highlight that the scrutiny of interviews conducted by the police in Australia, has not reached the ‘same level of intensity as the UK’. ⁵¹ Notably, the present authors found that Australian police forces are less transparent about procedures or questioning methods than their counterparts in the UK or NZ; Australian police do not disclose much information about the way suspects are interviewed.

In the absence of uniform legislation, implementation or transparency in procedural practice, the authors analysed the police manuals of the different Australian states which are publicly available. By analysing these publicly available documents, the authors aimed to provide an overview of the incorporation of investigative interviewing in Australia. ⁵² The authors discovered a formal endorsement of the PEACE model in two states (New South Wales (NSW) and Queensland). In other states and territories, there was no specific or implied endorsement of investigative interviewing, but its incorporation could be inferred by the research outlined below or other relevant resources such as training manuals or newsletters.

Concluding remarks

As discussed in this article, false confessions are a problem in Australia. One of the contributing factors may be interrogative police interview techniques. Hence, the reform of police interview techniques and practices is necessary to minimise the risk of false confessions. Literature suggests that investigative interviewing, and therefore methods to minimise the risk of false confessions, have been endorsed Australia-wide. The evaluation of the incorporation of the PEACE model in Australian jurisdictions found that there is reference and/or endorsement towards the investigative interview model in most of the assessed manuals. However, the practice, training and utilisation of the investigative interview strategy in each jurisdiction is unclear due to lack of public availability of resources such as training manuals, or procedures relating to conduct in police interviews. The authors understand and accept that, for security reasons, it is not possible for all police manuals and policies to be made available to the public.

Surprisingly, in the absence of formal documentation – unlike, for instance, the UK and NZ – Australia has only a few empirical studies that could indicate a change of questioning methods used during police interviews with suspects. Research in Australia lags behind compared to research in other countries, and the lack of transparency limits the opportunity to actively assess what happens during custodial interviews. ⁸⁵ The perception studies on police attitude are contributory; however, we do not know how police officers are conducting interviews with suspects as there has been only one empirical study on this topic, and this was in NSW some 12 years ago.

Despite this, the research discussed in this article is a first step towards assessing the effective incorporation of investigative interviewing in Australia. More research is needed to ensure the use of proper interviewing techniques nationwide, and their implementation in the interests of justice.