Introduction
The polygraph – or ‘lie detector’ as it is often knownFootnote 1 – is not fictional, as some might believe, nor is its use limited to reality TV shows or national security bodies in Russia or the US. Polygraph ‘testing’ is a recently established, yet opaque, practice in criminal justice in the United Kingdom (specifically, for the purposes of this paper, England and Wales), and its remit is continuously but covertly expanding. Previously limited in statute to certain sex offenders released on licence,Footnote 2 polygraph testing can now also be imposed on released domestic violenceFootnote 3 and terrorism offenders,Footnote 4 those subject to terrorism prevention and investigatory measures (TPIMs),Footnote 5 as well as those subject to sexual harm prevention orders (SHPOs) via a recent statutory power to impose positive requirements.Footnote 6 It is possible that the above-mentioned statutory measures are the only ones in English law that embed the training, qualification and operational requirements of an industry association from outside the UK (the American Polygraph Association (APA), which represents the polygraph industry in the US) into the justice system in England and Wales.Footnote 7
Until recently, the creeping expansion of the polygraph into criminal justice contexts in England and Wales has attracted little attention or concern from the wider public, Parliamentarians, the media or even from civil society groups, although a recent survey found that only 29% of respondents were comfortable with police use of biometric data to determine if someone was telling the truth.Footnote 8 We might speculate that the deployment of polygraph interviews in the context of sex, domestic abuse or terrorism offenders, is taking advantage of the weakened interest on behalf of the public as regards morally reprehensible, indeed serious crimes, or that the use immunity enshrined in section 30(1) of the Offender Management Act 2007 (OMA 2007)Footnote 9 would successfully mitigate any concerns.
(a) Focus on non-statutory use
This paper is concerned with uses of the polygraph by the police in England and Wales falling outside the above statutory regime. It might be assumed that statute law would restrict the mandate for the use of the polygraph by police forces within strictly circumscribed boundaries, particularly bearing in mind the National Policing Position Statement (2014), which strongly discouraged the use of the polygraph interview in investigations and security screening processes. This statement explained that ‘its use as a method of detecting deceit is at best a controversial one among psychologists’, thus raising ‘adverse consequences for the investigative interview, the wider investigation and the trial process’.Footnote 10 The position statement cited the British Psychological Society and the latter’s clear warning:
Anybody who attempts to persuade others (eg a polygraph examiner testing a police suspect) that polygraphic lie detection is an error-free procedure (by, for example, using a ‘stimulation procedure’ – Bull, 1988; National Research Council, 2003) will be attempting to deceiveFootnote 11
This is unsurprising, in view of the fact that even one of the main figures in the history of the polygraph, Leonarde Keeler, famously said that there was no such thing as a ‘lie detector’.Footnote 12
However, despite these warnings, our findings, based on FOI requests to police forces and government departments discussed further below, reveal opaque uses of the polygraph by the police that are not regulated by statutory provisions or other mechanisms. These include, as far as the data shows, the use of polygraph interviewing in connection with:
-
• suspects for child-protection decisions;
-
• community sentences;
-
• cautions, with polygraph testing being one of the conditions;
-
• risk-assessment for deletion from the sex-offender register;
In addition, there are restrictions around access to legal representation by those being tested.
(b) Foreseeability as framework of analysis
In this paper, we use an innovative approach combining doctrinal, law-in-context and empirical research elements, including analysis of European Court of Human Rights (ECtHR) case law and the Court of Appeal decision in David,Footnote 13 to argue that the legal basis for such polygraph interviews is inadequate and imprecise; that satisfying the test for necessity and proportionality would (or should) be a high(er) burden; and that lessons have not been learned from Bridges Footnote 14 or the cases that came before. We shall argue that the purposes for which police forces deploy the polygraph lack clear delineation and are unconstrained by independent oversight or statutory protections – including those in PACE, a key legislative framework intended to regulate police powers and protect individual rights. Our research suggests that without mandatory transparency and candour regarding the extent of the use of the polygraph by the police, it is difficult to see how a key principle underpinning human rights – foreseeability– can protect citizens from the risk of abuse. It has been argued that, since the Bridges case, the police have been afforded too much discretion to extend the remit of biometric surveillance.Footnote 15 In this paper, we show that – in a similar way – police forces are pushing their own boundaries regarding the deployment of the polygraph, at the same time deflating or ignoring the thresholds for legality, necessity and proportionality.
We have grounded our analysis in foreseeability and proportionality as underpinning principles of the human rights framework. Foreseeability is a fundamental part of the principle of legality, designed to minimise arbitrary and abusive application of the law by the state with the requirement for appropriate safeguards, and therefore uphold justice and fairness. The proportionality principle is at the core of the assessment of the reasonableness of a measure that restricts individual rights, which the polygraph significantly does. Analysis of proportionality implications is therefore vital, alongside a foreseeability assessment.
(c) Author stance
This paper focuses upon the non-statutory uses of the polygraph by the police and legal issues arising from this. The considerable uncertainty around the scientific validity of polygraph interviewing contributes directly to these legal issues, as we explain below. We recognise that some within the criminal justice community have found polygraph interviewing to have utility and value for risk management and facilitating disclosures. While this paper acknowledges and discusses this point of view, it should be noted that the authors do not find these views unproblematic.
(d) Structure of the paper
The remainder of the paper is structured as follows. Section 1 summarises the scientific position regarding the polygraph machine. Section 2 then reviews the legal basis for police use of the polygraph and analyses the extent of such use in England and Wales, and the applicable policies, as revealed by responses to our FOI requests. Section 3 discusses the implications of the remit of PACE, the issue of the voluntariness of the polygraph examination, and the availability or otherwise of legal representation, with Section 4 suggesting significant implications for the key principles of foreseeability, necessity and proportionality. In Section 5, we conclude that further attention must be paid to transparency and candour, and urge that it is time to abandon reliance upon the common law in respect of police use of contentious technologies and instead consideration be given to the introduction of a new ‘police law for technology use’.
1. The science behind the polygraph
Managing to separate true from false statements has been a dream of humankind since ancient times. The aspiration to detect lies in a reliable (ie scientific) way, has shaped mythology, literature, and science fiction; from Wonder Woman’s lasso of truth to Mary Poppin’s lie-detecting thermometer to the modern polygraph and ‘emotion’ AI, people instinctively strive towards developing a method of looking into the souls of others.Footnote 16
Crucial for the understanding of the polygraph is the premise that ‘deception and truthfulness reliably elicit different psychological states across examinees’Footnote 17 and in particular the belief that the polygraph is a device that measures certain ‘physiological responses … which are interpreted to determine whether an individual has responded to questions in a truthful or deceptive manner’.Footnote 18 However, the science is disputed; there is little to no mainstream scientific acceptance of correlation between physiological responses and the detection of lies. Ever since the first deployment of the polygraph criminal courts,Footnote 19 scientific institutions,Footnote 20 and military organisationsFootnote 21 have continuously and almost unanimously discredited the polygraph as regards its validity in fact-finding processes. What is more, the very scientific paradigm in psychology which propelled the polygraph into existence has receded due to its lack of methodology, indefensible empirical basis and thus deficient validity.Footnote 22 As the Royal Society notes summarily:
The polygraph is a well-known approach to detecting deception … it relies on the measurement of skin conductance, which can be influenced by arousal during deception – it has been repeatedly evaluated and its validity and reliability have been challenged for decades in systematic reviews and evaluations. In addition to questions about its reliability and validity, the polygraph is particularly vulnerable to countermeasures – covert or overt measures taken by the subject of the polygraph in order to distort or undermine any conclusions.Footnote 23
In other words, there is simply no unique physiological indicator that reflects a single underlying psychological process, let alone deception.Footnote 24 The realisation that there is no known physiological response which would be unique to deception, or indeed to any other cognitive state, pushes a sharp needle into the theory underpinning the polygraph.Footnote 25 What is more, even seasoned polygraph operators admit that the polygraph ‘test’ is primarily ‘a psychological procedure and only secondarily of a physiological nature’.Footnote 26 The emphasis on utility over validity is also reflected in the findings of the US National Research Council, when it states that one of the roles of the polygraph test ‘is to influence the conduct of a polygraph interview’. A polygraph interviewer, it observes:
who detects what he or she believes to be deceptive responses during the polygraph test normally conducts the remainder of the interview differently than an examiner who sees no signs of deception. Such an examiner may ask more probing questions, do additional charting, shift to a different type of polygraph test protocol, or take a more confrontational attitude in the interview in an effort to elicit an admission or to ‘clear’ the examinee of suspicion.Footnote 27
This means that according to its own terms, the polygraph is not an objective scientific procedure but an interrogation tool.
(a) The utility of the polygraph
Despite this unscientific nature, the use of the polygraph in the criminal justice system in England and Wales is justified by its proponents based on two main arguments. First, the utility argument: that even if the polygraph does not ‘work’ as a lie detector, its use is justified because more disclosures (confessions) are obtained from offenders. Secondly, the containment argument: that the protections and safeguards applied to criminal evidence are irrelevant to polygraph testing in a risk assessment, probation, investigatory or intelligence context.Footnote 28
The utilitarian approach to the polygraph is epitomised by the evaluation carried out after the passing of the OMA 2007. This study considered whether there had been more ‘clinically significant disclosures’ in an offender group subject to polygraph testing compared to a comparison group, focusing on numbers of disclosures rather than their truth-conducive character.Footnote 29 Furthermore, the 2023 review of counter-terrorism polygraph tests was a ‘process evaluation’ only, concluding that polygraph examinations had ‘embedded well into everyday practice’ and were regarded as an ‘effective risk management tool’. However, it was possible to detect some concerns among the positive messages – a finding of significant response where the person on probation was not forthcoming with information about why this might have occurred, and subsequent investigations found no information to corroborate the significant response, ‘sometimes … made the relationship with the person on probation difficult and impacted risk management’. Some counter-terrorism police and psychologists felt further consideration was needed around ‘proportionate responses, polygraph examination processes, eligibility, and the role of psychologists in the polygraph process’.Footnote 30
2. The policing polygraph landscape as revealed by FOI requests
(a) The legal basis for police use of the polygraph
Police forces in England and Wales operate based upon a tradition of ‘policing by consent’ built on the so-called ‘Peelian principles’. The rationale underpinning those principles is that the authority of the police relies upon the consent of the public and is not determined by an exhaustively codified regime of powers and duties.Footnote 31 Absent applicable statutory powers or restrictions, police powers stem from the common law; in a country ‘where everything is permitted except what is expressly forbidden’, the police can act lawfully, provided that there is nothing to make it illegal.Footnote 32 The police have common law power to obtain and store information for policing purposes, for the maintenance of public order and the prevention and detection of crime, including the overt taking of photographs.Footnote 33 Correspondingly, police constables owe the public a common law duty to prevent and detect crime.Footnote 34
However, this ‘citizen-in-uniform’, common law approach has constraints and limitations. The ECtHR has consistently held that a given measure relating to interference with rights cannot be regarded as valid law unless it is formulated with sufficient precision to enable citizens to regulate their conduct and to be able to foresee the consequences of a given action or situation.Footnote 35 Proportionality and the use of procedural guarantees against arbitrariness are essential in that regard.Footnote 36 Crucially, the requirement for sufficiently clear rules concerns both the circumstances in, and the conditions on which, the respective measure is carried out. If the court finds that there is a considerable risk of arbitrariness, the respective measure or even domestic law will not be compatible with the ‘lawfulness requirement’,Footnote 37 in view of the higher status of rights guaranteed by the European Convention on Human Rights (ECHR) – lex superior derogat legi inferiori.
So how does the legal framework around the use of the polygraph, especially as regards its remit, fare? An answer to this question presupposes a detailed overview of the polygraph landscape, including the scope and purposes of use, and policies detailing who will be subject to testing, how and in what circumstances. In view of the lack of publicly accessible information, further data was gathered through FOI requests.
(b) Non-statutory use of the polygraph by the police in England and Wales – analysis of FOI responses
We have taken a fact-finding approach to our analysis of the legal framework surrounding the use of the polygraph, including consideration of the policy and other extra-legal factors that may affect the effectiveness of the law. More specifically, we investigated the extent of non-statutory uses of the polygraph by police forces in England and Wales. FOI requests have been fundamental to this approach. Our FOI requests asked for information relating to the polygraph testing by police forces of:
-
• individuals suspected of, or charged with, an offence but not yet convicted, or who were under investigation;
-
• polygraph testing for child protection reasons;
-
• tests in connection with bail or community sentences; and
-
• testing for reasons other than for the management of sexual offenders, terrorism offenders, domestic abuse offenders and those subject to TPIMs.
(c) Comments on the FOI request method
Deploying FOI requests comes with challenges: drafting effective requests that will not be misunderstood or rejected; the likelihood that information will be withheld due to the availability of exemptions; and the challenge of comparing differing formats, styles of response and documentation. However, as researchers who in this context are ‘outside outsiders’,Footnote 38 the advantages outweigh the challenges. Our ‘round-robin’ FOI requests not only enabled us to obtain information about the extent of polygraph use which was not well known or publicly available; the responses also allowed us to identify inconsistencies in policy and practice between police forces, particularly valuable for research concerned with the consistent and fair use of state power and activities that raise rule of law concerns. As Savage and Hyde explain:
Freedom of information requests can be dispatched to multiple … authorities at the same time, allowing information held by public authorities to be obtained cheaply and in a uniform fashion. This data can then be usefully compared, allowing trends to be discerned.Footnote 39
In total, 91% of the police forces which were contacted (43) responded; 95% of respondent police forces included a statement neither confirming nor denying (NCND) whether any additional information relevant to the FOI request was held, by virtue of FOIA 2000, sections 23, 24, 30 and 31.Footnote 40 Interestingly, it was confirmed that this statement had been drafted centrally by the National Police Chiefs’ Council data protection and FOI central referral unit, and was justified by concern over revealing the intelligence or operational picture, or gaps in capability. What is more, the NCND statement was used inconsistently by police forces. For example, some forces applied it only to questions 1–2 of our FOI request (see Figure 1 below), another applied a partial NCND to a combination of questions, and another applied a full NCND. Some forces stated that no information was held regarding the use of polygraphs in respect of sex offenders and domestic abuse offenders. This inconsistency, combined with the differing responses mentioned above to investigatory uses, might allow certain inferences to be made, although we do not attempt to do so in this paper. It is certainly the case, however, that the application of NCND to the responses adds weight to the concerns about the opacity of police polygraph use, as discussed further in section 5(a) below.
Summary of FOI responses.
Four forces failed to respond, including, significantly, Greater Manchester Police (although a late response revealing a number of tests, but no policies, has now been receivedFootnote 41). Of those responding, 15% (ie seven forces) disclosed carrying out polygraph tests in circumstances mentioned above, and 21% (nine forces) disclosed policies, forms or other documentation related to such testing. Notably, a few forces denied using polygraphs at all. For example, Gwent Police stated that: ‘Polygraph is a specific form of integrity testing which we do not employ’.
(d) Purposes of polygraph interviews
Information about polygraph interviews carried out within the last six years was disclosed by seven forces. The aggregate number of such interviews for the purposes specified in our questions was as follows:
Furthermore, documentation disclosed pursuant to our FOI requests included (in no particular order): a polygraph pre-conviction guidance document (Suffolk); different versions of a voluntary polygraph testing information sheet/information sheet for risk management and assessment of post-conviction sexual offenders (Kent, Bedfordshire); a ‘voluntary testing consent’ form (South Yorkshire); a conditional cautions information sheet (West Yorkshire, Essex); a consent form in respect of conditional cautions (Suffolk); Standard Operating Procedure (SOP) Title: Polygraph for Post-Convicted Sex Offender Management (version 1, March 2022) – Redacted (Hertfordshire and Bedfordshire); National Polygraph Policy v12 ongoing, updated 14.03.23 (working Polygraph Operational Procedures) – Redacted (Essex); Polygraph Examinations Procedure – Northumbria Police October 2021; National Offender Management Service – Mandatory Polygraph Testing: Information for Offenders (Ministry of Justice); Ministry of Justice Polygraph Examination Framework;Footnote 42 ACPO May 2014 National Policing Position Statement: the Use of the Polygraph in Investigations (as above); a link to software manuals provided by the US-based commercial provider (which we understand is the sole provider) of the polygraph software algorithm and hardware (Lafayette Instrument)Footnote 43 (Ministry of Justice, North Yorkshire, Cumbria); a link to the ‘Utah Numerical Scoring System’ (via an article from 1999, Home Office, Ministry of Justice).Footnote 44
The above documentation and the FOI responses indicate that (in addition to testing pursuant to an SHPO) police forces may or could deploy polygraph testing for the following purposes:Footnote 45
-
• ‘voluntary’ risk assessment of post-conviction sex offenders (all offenders subject to part 2 of the Sexual Offences Act 2003) but not the statutory assessment of licence conditions conducted by the Probation Service (SOP 2022);
-
• assessment of offenders who apply for removal from the sex offenders’ register (SOP 2022);
-
• assessment of offenders within six months of such removal (SOP 2022),
-
• assessment of ‘any other offender for whom the offender manager deems that polygraph examination may assist in the effective risk management of that offender’ (SOP 2022);
-
• in connection with review of indefinite notification requirements ‘as a means of demonstrating their attitude to offending and reassuring the police that their risk has now been minimised’ and in relation to other registered sex offender discharge applications to ‘display that their risk is now minimised’ (Northumbria Examinations Procedure);
-
• the issuing of conditional cautions with a ‘requirement’ to take part in a polygraph examination as a means of disposal for the offence of failing to comply with sex offender registration requirements, breach of SHPO, ‘or other offences if the criteria are met’, where non-cooperation is deemed to be a failure to engage and could result in prosecution for the original breach (SOP 2022, Suffolk Constabulary consent form);
-
• risk management of those suspected of committing sexual and ‘violent offences’ (Northumbria Examinations Procedure);
-
• testing of ‘those who have been arrested on suspicion of IIOC [indecent images of children] offences’ (Suffolk);
-
• ‘enhanced risk assessment of post-conviction sexual offenders, enabling appropriate offender management and safeguarding of potential victims of sexual offending through use of polygraph’ (Bedfordshire Police);
-
• testing of ‘suspects of online child sexual abuse offences’ (2023 Operational Procedures);
-
• in connection with child contact, ‘A truthful test may be used to support decisions regarding child contact’ (Kent);
-
• in connection with bail decisions/community sentences or orders (disclosed by Essex);
-
• in relation to ‘domestic abuse’ (2023 Operational Procedures);
-
• in relation to ‘child sex offender disclosure scheme’ (2023 Operational Procedures);
-
• in connection with Multi-agency Public Protection Arrangements (2023 Operational Procedures);
-
• in connection with ‘stalking protection orders’ (2023 Operational Procedures);
-
• in connection with three further uses redacted from the 2023 Operational Procedures.
Many of the above purposes relate to risk management or risk assessment of those convicted or suspected of sexual offences, and the facilitation of admissions or intelligence disclosures. The 2023 Working Polygraph Operational Procedures state that ‘any use outside these agreed parameters should be agreed through the NPWG [National Polygraph Working Group] and in matters of urgency through the Chairperson’. It is therefore possible that other uses, undisclosed or redacted, may exist. One police force expressed an ambition to ‘embed polygraph pre-con [pre-conviction] testing within various other departments where applicable, so [that they] can safeguard as many victims and vulnerable people as possible’.Footnote 46
Furthermore, College of Policing minutesFootnote 47 note that ‘one force recently used polygraph evidence in a statement’, contrary to the above-mentioned ACPO position statement which strongly discourages use of polygraph examinations in investigations.Footnote 48 We should note, finally, that in response to our FOI request, two forces stated that they do not ‘routinely’ use polygraph tests for investigatory purposes (Bedfordshire and Cambridgeshire), while another two forces said that they do not use testing for investigatory purposes for sex and domestic abuse offenders (Devon and Cornwall and Dorset). Yet another disclosed that they do not deploy ‘overt’ testing (Gloucestershire): due to the ‘neither confirm nor deny’ proviso deployed by all forces, it was not possible to ascertain the extent (if any) or nature of ‘covert’ testing.
3. PACE, voluntariness of polygraph examination, and legal representation
(a) The remit of PACE 1984
In all three of the procedures disclosed by police forces (as mentioned above), it was admitted that ‘[t]here is currently no UK law governing the use of the polygraph by police for risk assessment and management of sexual offenders’.Footnote 49 Indeed, what legal basis justifies the taking of child contact decisions (which may include contact with the individual’s own children) on the basis of polygraph examination, for instance? It appears that, in a similar way to the purported legal basis for the deployment of live facial recognition, police forces are relying on a patchwork of uncoordinated fragments from different areas of law, notably common law and general statute, to justify the extraction of information via the deployment of a polygraph interrogation tool, and the taking of decisions based on polygraph interview results. This should be contrasted to the clear statutory provisions governing polygraph interviews by the Probation Service in the context of – we cannot stress this enough – a closed number of offences.
Nevertheless, the polygraph operating procedures assert that ‘polygraph examination will be conducted with regard to the provisions of the Police and Criminal Evidence Act (1984) and Codes of Practice and the Human Rights Act’.Footnote 50 Worryingly though, PACE does not govern the use of polygraph interviews in any interrogation context: indeed, it contains no mention thereof. In contrast, the UN Manual on Investigative Interviewing for Criminal Investigation, which advises on human-rights compliant interviewing based on the Mendez Principles on Investigative Interviewing and Information Gathering (2021),Footnote 51 states that ‘[t]he use of lie-detection technologies is ineffective and may lead to errors in justice’.Footnote 52 What is more, documentation disclosed suggests an inconsistent approach to the application of PACE. Despite the importance of independent legal representation of suspects and offenders (following conviction), a common and dominant theme in the documentation was the exclusion of such representatives from the polygraph interview – police forces stated that no legal representative or third party is to be permitted to accompany the interviewee. There appeared to be no exception made for vulnerable interviewees. Nor would free legal advice be provided, a decision justified by the assertion that ‘[a] polygraph examination is not a criminal interview conducted under PACE 1984’Footnote 53 – which contradictorily, runs counter to the praying-in-aid of PACE as a legal framework for polygraph interviews.
Procedural fairness thus appears to be a significant concern, bearing in mind the semi-evidential and semi-investigatory nature of several of the purposes (see above section 2b) for which polygraph interviews are deployed by the police, including for decisions relating to the sex offenders register, conditional cautions in which polygraph testing is stated to be mandatory, in connection with stalking protection orders and, importantly, the interviewing of suspects for decisions relating to child contact where a parallel offence is being investigated. It could be argued that polygraph interviews are not being used in a conventional ‘criminal investigation’ or ‘evidential’ context. However, polygraph interviewing or agreement to such interview in many of the circumstances mentioned above appears to link to a high-stakes decision being taken about an individual facing significant consequences – such as issuing of stalking protection orders, receiving a conditional caution versus prosecution – potentially without the individual having access to independent advice to weigh up available choices. It has not been disclosed whether a police caution as specified by PACE is issued to the interviewee before such deployment of the polygraph.
(b) On voluntariness and legal representation
As mentioned above, the disclosed documentation asserts that the polygraph interview is ‘entirely voluntary’, and not ‘a criminal interview’. Therefore, they add, ‘free and independent legal advice is not available’; speaking to a solicitor should be done before the subject attends the polygraph interview and is ‘at their own expense’.Footnote 54 This suggests an assumption that ‘voluntariness’ will supposedly remedy the polygraph interview’s extraction from the remit of PACE. Of course, an interview can always be said to be voluntary, to the extent that it would be unthinkable to (physically) force someone to undertake the polygraph test, but the question of ‘voluntariness’ does not ride on physical coercion. In a case involving the question of false imprisonment and enforced compliance, the Supreme Court held that even though it was physically possible for the claimant in the case to break his curfew, his compliance with the restrictions was enforced and not voluntary, due to threats of force, legal process or further confinement, backed up by the power of the state.Footnote 55
Indeed Code C (detection, treatment and questioning of persons by police officers) of PACE itself recognises that ‘[t]he rights, entitlements and safeguards that apply to the conduct and recording of interviews with suspects are not diminished simply because the interview is arranged on a voluntary basis’.Footnote 56 Relatedly, in the ECtHR cases of Wang v France Footnote 57 and Dubois v France,Footnote 58 both applicants had consented to being interviewed and had been informed of their right to end the interview at any point, in accordance with the law as applicable at that time. However, they had not been expressly informed of their right to remain silent and had not been offered an opportunity to obtain legal assistance and, in one case, the assistance of an interpreter.Footnote 59 The Court found, with regard to the defence rights protected by Article 6(1) and (3)(c) of the ECHR (right to a fair trial), that the requisite safeguards in the context of a voluntary interview should be the same as those applicable to police custody.
(i) Both inside and outside of PACE
It is beyond the scope of this paper to conclude definitively whether each of the purposes for which polygraph examination is deployed by the police in England and Wales falls within the scope of Article 6 of the ECHR. It is not engaged, for instance, where polygraph interviews are imposed pursuant to an SHPO as a ‘safeguarding’ measure.Footnote 60 However, a polygraph interview is an interrogation tool designed to generate adverse statements. As the National Polygraph Examination Policy (Standard Operating Procedure) states: ‘The focus of the polygraph process is not solely on whether the subject shows significant responses on a test, but on facilitating admissions or intelligence disclosures.’Footnote 61 In the probation context, such disclosures are claimed to assist in the assessment and supervision of offenders for risk management purposes. The above policing purposes potentially extend beyond risk management, influencing decisions relating to an individual’s treatment by the criminal justice system, including determination of an application for removal from the sex offender’s register, and whether an individual receives a conditional caution as opposed to facing prosecution.
Further, although polygraph testing relating to decisions over child contact could be categorised as ‘safeguarding’, there appears to be a close linkage to parallel investigations into potential offences relating to indecent images of children. It must be questionable, therefore, whether polygraph examination relating to child contact can in reality be separated from such parallel investigation. The content of the 2023 procedure suggests that there is certainly a blurred line: the polygraph procedure will be part of ‘disclosure’ but not part of ‘criminal proceedings’; it will be included in the ‘MG6D’ (police schedule of relevant sensitive material for the prosecutor) but polygraph chart data will be excluded from subject access requests; the information is ‘intelligence’ only but ‘early investigative advice may be sought where disclosures from a polygraph examination form part of the investigative strategy’.Footnote 62
So, despite the rights-impacting consequences of police polygraph examinations, the polygraph is said at the same time to be within and also outside the scope of PACE and other disclosure, investigatory and intelligence processes and procedures. PACE, however, is a coherent web of safeguards in respect of the powers and duties of the police. Cherry-picking aspects of PACE severely disrupts its procedural architecture and protection.
(ii) Voluntariness and polygraph as ‘lie detector’
What is more, it is highly questionable whether the interview could be said to be voluntary in any of the circumstances covered by the policies disclosed. As we explored above, the polygraph interview is designed as an interrogation tool. Unlike door-to-door enquiries or voluntary interviews of witnesses, refusal to take a test and admissions made by the individual during that process, can have direct and serious consequences for the interviewee. Bearing in mind the steeply asymmetric imbalance of power between the citizen and the state, it would be a matter of priority to determine the impact on the relationship between the police and the individual of a refusal to volunteer: for instance, how is such refusal recorded and what impact, if any, does this have on the future classification and monitoring of that individual? Neither the responses to our FOI requests nor (as far as we are aware) other publicly available information provide answers to these questions.
What is more, the supposedly voluntary character of the polygraph interview should be viewed in light of real-world contexts. If the subject is ‘found to be truthful’, then, the Kent Police information sheet adds:
‘a) [his/her] risk could lower.
b) If the subject was applying to come off the register after 15yrs, a truthful test may support [his/her] application.
c) a truthful test may be used to support decisions regarding child contact’.
A positive outcome is utilised as strong incentive for ‘voluntary’ participation. Such an outcome, however, is merely the flipside of a negative one. The suspect must make a decision under pressure. The information sheet further states that the subject will ‘fail’, ‘if the results suggest that [the subject has] not told the truth’, suggesting that the polygraph is being presented to the subject as able to detect lies, thus potentially increasing the pressure for ‘voluntary’ participation. This is not only contrary to scientific consensus but does not cohere with the Ministry of Justice Framework, which advises that:
The term ‘lie detector’ should not be used either verbally with the individual or in any written form. (para 2.6).Footnote 63
This presentation of the polygraph as ‘lie detector’ is a crucial aspect of the question of voluntariness. Where it is suggested that polygraph testing can determine truthfulness, the portrayal of the interview as ‘voluntary’, in circumstances critical to the individual, is disingenuous at best, and emphasises the importance of the PACE contradictions and exclusion of legal advice discussed above.
(iii) Oppression concerns
Furthermore, there are a number of references to section 76(2)(b) of PACE within the documentation disclosed, and discussion of the risk that confessions made as part of the polygraph procedure, if used in evidence, are likely to be challenged on the grounds of oppression or unfairness (importantly, courts in England and Wales accept that the use of deliberate deception on a suspect may contribute to a finding of oppressionFootnote 64). Relatedly, it is stated that confessions will be used as intelligence ‘but may lead to further investigation with the purpose of initiating criminal proceedings’ (such as interview under caution or arrest).Footnote 65 It appears, therefore, that the polygraph test may be used as an interrogation tool in order to extract information leading to further investigation (notwithstanding the acknowledged ‘oppression’ concerns), as well as for the semi-evidential/semi-investigatory purposes mentioned above, but without the interview or information acquisition safeguards under PACE or elsewhere. This should be a reason to be concerned.
4. Forseeability of relevant law
The Ministry of Justice Polygraph Examination Framework notes that ‘polygraphing an individual outside the legislative and policy framework may breach their Article 8 [ECHR] right to privacy’.Footnote 66 Yet, for police use of the polygraph, it is admitted that there is no specific UK law; rather, polygraph examinations will be conducted with regard to PACE and the Human Rights Act 1998. As discussed above, however, there appears to be little consistency or coherence in the application of these frameworks, with the testing process stated to be both within and outside PACE safeguards, and policies containing no clear boundaries around the extent of deployment.
(a) Foreseeability
As we briefly explained above, a legal rule authorising police operations needs to be both accessible and foreseeable to prevent arbitrariness in the application of the law. As the Strasbourg Court notes, domestic law must be sufficiently clear to give citizens an adequate indication as to the circumstances in which, and the conditions on which, public authorities are empowered to resort to any such measures.Footnote 67 Moreover, the law must indicate the scope of any discretion conferred on the competent authorities and the manner of its exercise with sufficient clarity to give the individual adequate protection against arbitrary interference.Footnote 68
It is not accidental that the twin requirements of accessibility and foreseeability have consistently featured in case law under Articles 5, 6, 8 (but also 10) of the ECHR. It is well-settled human rights law that the respective court must carry out a close examination of the procedural safeguards embedded in the system to prevent arbitrary encroachment on individual rights.Footnote 69 To summarise: the ECtHR has consistently held that a measure cannot be regarded as valid law, unless it is formulated with sufficient precision. This will enable citizens to regulate their conduct, and to foresee – if necessary with appropriate legal advice – the consequences of their actions.Footnote 70
The question, therefore, is what mechanisms have been put in place to minimise arbitrariness to the extent that this is possible. The answer cannot be given in a procedural vacuum. The Court of Appeal in Bridges accepted that the more intrusive the act, the more precise and specific must be the law said to justify it.Footnote 71 There would appear to be little reason to doubt that subjecting a person to a polygraph interview, an interrogation tool designed to extract adverse statements, is likely to be regarded by the courts as sitting at the higher end of intrusiveness, bearing in mind the bodily interference and discomfortFootnote 72 involved in the interview, the sensitive biometric data collected, the long, choreographed questioning processes, as well as the opaque and incontestable conclusions about deceptiveness drawn from the process (the validity of which is itself contested).
It might be argued that polygraph interviewing of offenders, or even suspects, sits on a comparatively lower scale than, say, the collection and retention of DNA samples of those who were suspects of crime but not convictedFootnote 73 or the scanning of ‘innocent’ crowds by a live facial recognition tool.Footnote 74 The fact remains, however, that the polygraph is a physically intrusive, unscientific (and, thus, of questionable necessity), technique involving the collection and analysis of sensitive biometric data, and furthermore (as we discussed above) raises far more fundamental issues of scientific validity than facial recognition. Facial recognition ‘works’ even though it may often generate (measurable) errors. Of course, everything – including the polygraph – is relational, and intrusiveness might, under different circumstances, be offset by the tool’s reliability or utility. But no impartial stakeholder, not even the UK Government, asserts that the polygraph is based on valid science. The very existence of OMA 2007, section 30,Footnote 75 and of the use immunity enshrined therein, presupposes the inadmissibility of polygraphic evidence.
Although (as discussed in Bridges) local policies could constitute part of the relevant law, as far as we are aware, the documentation disclosed via the FOI responses is not widely circulated or published, nor does it suggest a consistent approach across police forces in England and Wales as regards the circumstances in, and conditions under which, the polygraph will be deployed. There appears to be a significant risk that police conduct, as per the policy and guidance documents, may not satisfy the ‘quality of law’ requirement, as being insufficiently precise. It is anything but clear who will be subject to polygraph interviews, in which circumstances and subject to which safeguards. It is also unclear, indeed it is irrational, to assume that polygraph interviewers will draw rational – ie verifiable – conclusions. To use language deployed by the ECtHR, ‘[s]uch a lack of transparency, at the very least, can hardly meet the requirement of foreseeability, this in turn being one of the preconditions for the lawfulness of any interference with the rights protected by Article 8 of the Convention’.Footnote 76
Furthermore, common law cannot authorise ‘intrusive’ ways of obtaining information, such as entering private property, acts constituting assaultFootnote 77 or covert surveillance.Footnote 78 Purshouse and Campbell contend that ‘forms of “physical” intrusion, such as DNA sampling and bugging private property, fall beyond the scope of the common law powers of the police, thus requiring a more narrowly prescribed legal basis’.Footnote 79 Relying on the case of Collins,Footnote 80 they also suggest that ‘as well as physical intrusions, excessive demands to answer questions, including the use of actual or implicit threats, fall outside of the common law powers’.Footnote 81 The physically intrusive nature of the polygraph test, combined with the power-imbalance surrounding its deployment by the police, would likely remove polygraph interviews by the police from the remit of the common law.
(b) Polygraph methodology and implications for foreseeability
Polygraph subjects need to be able to know which methodology their interviewers will rely on to assess their physiological data. This goes to the heart of the problem, for such a methodology based on a reliable scientific basis is simply absent. As a result, neither citizens nor indeed anyone else can reliably predict either in advance or ex post facto how physiological data will be assessed and which conclusions regarding truthfulness should (not) be drawn.Footnote 82
Early on in the history of the polygraph it became clear that ‘wide divergence’ in the structure of the respective interview is inevitable due to the ‘widely varying types of questions, examiners, and examinees’.Footnote 83 The complex interaction between the examiner and the examinee show that lack of standardisation signals a feature, not a bug in the system. Although it is understandable in a policing context that each test is bespoke to the individual, what is lacking is a scientifically provable, reliable and repeatable method to determine whether a statement made is truth or lie, or even if a statement indicates deception. Ultimately, therefore, the term ‘test’ is a misnomer. The polygraph procedure is at best an interrogation technique, yet lacking in appropriate procedural protections, as discussed above. This creates the tangible risk of arbitrariness and lack of foreseeability, which according to the ECtHR is more than half the way to a violation of human rights. Proponents of polygraph testing may point to the references in policy to APA ‘standards’ and training requirements. Unlike other frameworks, such as forensic science regulation,Footnote 84 designed to ensure accurate and reliable scientific evidence and validation of methods, APA policiesFootnote 85 lay out processes developed by the polygraph industry itself, designed for a US context, absent the rigour of independent validation or assessment of techniques deployed.
The methodology and the rules on which scientific validity is based should, by definition, be articulated, scrutinised and capable of being rejected. Lacking clear operational procedure and rules of assessment for polygraphic data deprives both suspects/offenders (and their representatives) and polygraph interviewers or fact-finders of the ability to make an informed decision. A polygraph interviewer alas cannot follow any rule, for there is no standardised way to conduct an interview or assess the data. They cannot work out which inferences are to be drawn by consulting, say, an authoritative table showing the probative force of certain physiological patterns. Therefore, we continue to lack a reliable and replicable method accessible to polygraph examinees and their legal representatives.
(c) Necessity and proportionality
In a recent (2023) case, David,Footnote 86 the appellant, a serious sex offender, challenged a positive obligation in an SHPOFootnote 87 that required him to comply with any instruction from his offender manager or the police to attend polygraph/integrity screening and to comply with any instructions during those sessions by the person conducting the assessment. This requirement applied save where there was ‘good reason’ not to do so.
(i) Necessary, clear and effective
The Court considered previous case law relating to the reasonableness of requirements and restrictions imposed in SHPOs made under the Sentencing Act 2020. It noted in its judgment the case of Parsons,Footnote 88 which held: (a) that no order should be made unless necessary; and (b) if necessary, then the prohibitions imposed must be effective, clear and realistic. It was held in David that the judge imposing the SHPO had not stated why the polygraph interview requirement imposed was not only necessary but also proportionate and sufficiently clear, such statement being required by the statute. The SHPO was drafted in terms which were too wide and vague, and was therefore disproportionate and oppressive. The SHPO required the appellant for the rest of his life to take a polygraph test for any reason, or for no reason at all. The remit was not defined, and it was impossible for him to predict whether he had good reason to refuse to comply.Footnote 89 The Court also noted that the SHPO already contained a comprehensive set of measures designed to protect the public from sexual harm, raising questions as to why the polygraph testing requirement was necessary.
However, submissions made by the appellant in David, based on rights under Article 6 and Article 8 of the ECHR, were not considered to have any merit. The Court held that neither criminal nor civil rights were engaged, as the polygraph requirement was being used as a safeguarding tool, and in the context of prohibitions imposed pursuant to the Sentencing Act 2020, ‘any interference will be in accordance with law and justified as permitted by Article 8’.Footnote 90 This latter rather perfunctory assessment did not address the requirements of the accessibility and foreseeability of the law in any depth. Neither did the Court take account of the lack of specificity of the positive requirements in the amended Sentencing Act; the nature of those requirements as they relate to the polygraph is not mentioned in the statute.
(ii) Grounds for future proportionality challenge?
Note that David was decided in the context of statutory provisions mandating requirements in an SHPO to be ‘necessary’ to protect the public from harm. We cannot therefore merely read across the principles in the decision to any future necessity and proportionality challenge to non-statutory uses of polygraph testing by the police, potentially based on the four-part Bank Mellat test.Footnote 91 Murray distils this test into two main criteria: why is a deployment required; and what alternative mechanisms are available?Footnote 92 When determining necessity for rights interference in terms of the existence or otherwise of a ‘pressing social need’ for the measure in question, it is not merely a matter of demonstrating usefulness, relevance or concerns around security, cultural issues or public opinion.Footnote 93 The test requires an assessment of the alternative means that could be deployed and which may be less rights-infringing (as the judge pointed to in David), therefore involving an assessment of potential impact,Footnote 94 and a final balancing stage involving ‘normative evaluation and judgement’.Footnote 95 Despite the disparate concepts of proportionality found in English law,Footnote 96 we might reasonably predict that the issues of vagueness/clarity, the wide breadth of the power,Footnote 97 their effectiveness and in particular consequences of refusal highlighted in the David judgment would likely be crucial factors in any proportionality decision, as would the alternatives to polygraph testing available to achieve the desired objective and the existence (or otherwise) of independent safeguards or oversight.
Whilst a wide margin of appreciation is permitted by the courts in issues of public security, procedural safeguards remain crucial,Footnote 98 which – as demonstrated above – are lacking for non-statutory police uses of the polygraph. Furthermore, the state’s margin of appreciationFootnote 99 in respect of qualified rights is not unlimited where issues of individual vulnerability or personal and family life are in play.Footnote 100
Relatedly, the Court of Appeal in Bridges regarded the complainant’s brief interaction with the live facial recognition scanning as involving only ‘negligible’ impact.Footnote 101 The impact for polygraph interviewees cannot be described in a similar way. It is therefore arguably surprising that ‘effectiveness’, in terms of (lack of) scientific validity, likelihood of error and consequential impact on the proportionality assessment, was not addressed head-on by defence submissions in David. However, potential issues with access to independent scientific experts may mean that such arguments are more difficult for defence counsel to make.
(iii) Equality and bias
The discussion around scientific validity in Bridges was confined to issues of bias within the tool on sex and race grounds, and to consequent failure to comply with the public sector equality duty in the Equality Act 2010.Footnote 102 In comparison, discussion of issues of equalities and discrimination were absent from the documentation received in response to our FOI requests. Although the 2023 national polygraph examination policy stated that polygraph examiners were trained to carry out an assessment of mental capacity,Footnote 103 we found no indication in police information disclosed of the results of any equality impact assessments (EIA), nor any indication that polygraph interviewees had access to independent medical assessment in relation to any physical or mental conditions. In contrast, a 2023 EIAFootnote 104 relating to a Bill states that some people with neurodiverse needs would ‘not’ be suitable for testing, including those with memory loss or dementia, and caution would be needed for people with learning difficulties.
(iv) Oppressiveness and disproportionality
The issue of oppressivenessFootnote 105 – linked to disproportionality and which the Court found so important in David – also looms large in the documentation disclosed pursuant to our FOI requests. For the consequences of refusing a polygraph interview are potentially serious, indeed existential, for the individual. What reasons would be considered ‘good’ enough for a suspect or a person offered a conditional caution for instance to refuse to take a test? Could concerns over scientific validity or lack of information provided to advisers cross this threshold? In terms of information provided to the interviewee, Norfolk Constabulary told us:
We do not give the examinee anything in writing regarding the pre and post-test. These phases are not explained to the examinee other than informally telling them that we will be asking them questions around their health and welfare and then questions surrounding their family life, free time, employment, etc, to get to know them better. We then explain that we will go through the questions we intend to ask them and that these can be re-worded, deleted or exchanged for another question, if required … With regards to the post test, we explain prior to the test that once we have completed the test, it will be marked, and they will be given the result prior to them leaving. We will also give them an opportunity to explore any questions that they may fail on. We do not have any legal representation involved in the polygraph procedure. (emphasis added)
Subject access requests are likely to result in only limited information disclosure, as the 2023 Operational Procedures recommend that ‘the audio/visually recorded footage and polygraph chart data is exempt from all subject access requests to protect sensitive tactics’. From a data protection perspective, therefore, individuals and their advisers would be unable to review the accuracy and completeness of the output, inferences and conclusions generated by the testing process. Thus, both data protection and access to legal advice, as safeguards, lack teeth. This is compounded by the risk of oppressiveness by way of misleading claims made in some documentation about the polygraph’s truth-revealing qualities, thus leading the interviewee to believe – incorrectly – that the polygraph can detect lies and truths.
(v) Proportionality predetermination
Proportionality, as a legal test and standard, should ‘reach-back’ to shape the policy-making process and practical operations.Footnote 106 However, assessments of proportionality in ex ante impact assessments ‘might be shaped in a way to reinforce, rather than challenge, the starting assumptions that underpinned the initial policy formation’.Footnote 107 Fussey and Sandhu point to a tendency to view public safety as the ‘higher purpose’ over citizen’s rights, misconstructing the characterisation of proportionality ‘in terms of achieving policing aims rather than addressing the degree of rights interference’,Footnote 108 ie as a post facto rationalisation (attempt) of a decision influenced by political ideology.
The National Polygraph Examination Policy states (in respect of the Human Rights Act 1998): ‘The examination of sexual offenders and persons arrested/interviewed for offences of IIOC is done with the consent of the subject and is proportionate to what it seeks to achieve in terms of risk assessment’. In respect of necessity, the policy states: ‘Police must take reasonable action to protect the general public, particular individuals and groups, from serious harm and re-victimisation.’ The polygraph, it stresses, ‘is necessary to fully assess the risk a subject presents and take the appropriate action to address such risk’.Footnote 109 Both statements suggest a predetermined blanket conclusion of necessity and proportionality for the policing purposes covered by the policy, despite the polygraph’s lack of a sufficiently reliable scientific basis, and a risk, as Katwala succinctly puts it, of succumbing to a ‘natural instinct to give [a flawed technology] more weight than it deserves, or to use it as psychological prop or a propaganda weapon’.Footnote 110
5. Moving forward
(a) Transparency and candour
Clause 73 (‘Ethical policing (including duty of candour)’) of the now-abandoned Criminal Justice BillFootnote 111 would have required the College of Policing to issue a Code of Practice setting out actions for the purpose of securing that the police act ethically. Acting ethically was stated to include ‘acting in an open and transparent way in relation to the way in which the police have conducted themselves’, except where doing so would prejudice national security, prevention, detection, investigation or prosecution of any offence, or other public interests.
The opacity surrounding polygraph interviewing by the police, including the limited information provided to examinees about the testing process, the exclusion of legal representatives from the test, the de facto derogation from PACE, as well as the inconsistencies uncovered in the FOI responses themselves, are all factors pointing to the critical nature of a ‘duty of candour’.Footnote 112 An undertone of equivocation, rather than candour – potentially at odds with police forces’ duty of advice and assistance under the Freedom of Information ActFootnote 113 – might be discerned in the following extract from minutes of a National Polygraph Working Group, in which it was noted that forces were receiving FOI requests from a variety of sources:
FOI- every force receiving requests pertaining to Polygraph- always been a drip feed but received a lot of late-…taken out some personal and commercial matters- bits that are tactical and may undermine operational policing- more difficult to do so now- ICO changes etc- struggling to redact much at all- bear in mind with conduct of meetings going forward approach was sent to GC and I for coordinated approach- volume high- met with NPCC CRU agreed that they will instigate a national referral process-…some data has gone out previously- shut door after horse has bolted- difficult to hold some info back- CRU will be able to hold some matters back.Footnote 114
The protections offered in theory by the human rights and criminal justice frameworks will be stymied without an independently-enforced requirement for transparency and candour regarding the extent of polygraph testing within policing, the policies pursuant to which polygraph interviews are carried out, and the errors and risks involved in its use.
Conclusion
Our findings raise questions around the lawfulness and procedural fairness of the non-statutory regime described above, and add another dimension to the perennial question about the flexibility of police common law powers versus a codified regime of powers and rights, mandatory policies and codes. Many of the purposes for polygraph testing disclosed are both surprising and concerning, including high-stakes decisions around child contact and the imposition of polygraph interviews as a ‘mandatory’ requirement for receiving a conditional caution. Bearing in mind the contested scientific nature of polygraph testing, there is a serious risk of police actions being led in the wrong direction, thus undermining public trust in policing and the moral legitimacy of the law.
Despite the clear links between polygraph testing and the investigatory and intelligence-gathering processes, police deployment of the polygraph in risk assessment and intelligence contexts falls into a legal twilight zone, said to be both subject to PACE, yet also outside its remit, with interviewees denied legal representation during the interview itself. Legal protections are reduced, due to the deployment of polygraph testing in contexts which are described as risk assessment, ‘non-evidential’ or safeguarding; yet the lines between these activities and investigatory or evidential processes are blurred to say the least. The results of a polygraph examination can have life-changing consequences for the individuals involved. Without urgent clarity, or case law, we fear that police forces and governmental departments risk a landmark ‘abuse of power’ case, with compensatory, operational and reputational consequences.
We thus strongly urge the College of Policing, in view of its new remit, to establish a National Polygraph Training School,Footnote 115 to conduct a full and independent review of all (statutory and non-statutory) uses of polygraph interviews involving police forces. This review should cover the scientific validity of such uses and the processes surrounding them, including those redacted from responses to our FOI requests, and the full gamut of legal issues arising, not only those we have discussed in this paper but other relevant issues requiring further research and consideration, including privacy, reliability and duties of care. Notwithstanding, the documentation disclosed clearly indicates that at least some police forces (that we know of) intend – or have possibly already started – to deploy polygraph interviews for the investigation of non-sexual offences, bringing the polygraph ever closer to a standard operational process in policing. If an unscientific process such as the polygraph interview becomes mainstream within policing in England and Wales, what will the future hold? ‘Emotion AI’, perhaps, the premise for which is equally contested but arguably not physically intrusive?Footnote 116 A recent report from the Alan Turing Institute comments:
Current laws are failing to keep pace with changes to biometric technology, which risks undermining public confidence and trust in these systems. Most notably, the current legal framework does not adequately distinguish between tried and tested, scientifically valid biometric systems (such as fingerprint identification, DNA analysis and facial matching) and novel, often untested inferential or classificatory systems – such as age estimation, emotion recognition and gait analysis.Footnote 117 (emphasis added)
Intrusive methodsFootnote 118 in other sensitive contexts such as deployment by the intelligence agencies are subject to independent oversight and authorisation. Therefore, we conclude that it is time to abandon the reliance on the common law and a jigsaw of general statutes, and put in place ‘police law for technology’, covering risk assessment, non-evidential, intelligence and safeguarding purposes, as well as crime-investigation and evidential contexts, with genuine transparency and independent pre-approval processes to determine whether – and if so, how – the police should be permitted to deploy the polygraph and other contentious technologies.
Open access
