Normative Evolution in Evidentiary Exclusion: Coercion, Deception and the Right to a Fair Trial


Normative Evolution in Evidentiary Exclusion: Coercion, Deception and the Right to a Fair Trial



THE DISCIPLINARY FIELD traditionally known to common lawyers as the Law of Evidence is often regarded as essentially, if not exclusively, preoccupied with epistemic considerations. Nowadays the Law of Evidence is routinely associated with proof and truth, accurate fact-finding, convicting the guilty and acquitting the innocent. Bentham was the genius and pioneering advocate of an epistemic approach to English legal process, and his great project has been revived and extended by modern legal scholars, notably William Twining,1 and contemporary neo-Benthamites like the philosopher Larry Laudan.2 Echoes of Bentham’s insistence on rational fact-finding uncluttered by doctrinal baroque regularly issue from the mouths of experienced judges and apex courts, including the House of Lords3 and the UK Supreme Court.4 ‘Freedom of proof’ is the order of the day.5

From an epistemic perspective, fairness-based exclusionary rules immediately appear problematic because they authorise, or possibly even require, that ex hypothesi relevant information should be withheld from the trier of fact. This is prima facie irrational: in Bentham’s succinct aphorism, to exclude evidence is to exclude justice.6 There are essentially two strategies for reclaiming fairness-based exclusionary rules for orthodox evidentiary theory. First, one might demonstrate that fairness-based exclusions, on closer analysis, actually serve epistemic objectives, thereby dissolving their apparent irrationality. Alternatively, one might treat fairness-based exclusions as exceptional deviations from the law’s epistemological priorities, leaving the question of their contextual legitimacy to ad hoc rationalisations. This approach is reflected, for example, in Wigmore’s designation of ‘rules of extrinsic probative policy’7 and in Cross and Tapper’s expanding discussion of ‘judicial discretion’ as the slightly shady alter-ego of authentic rules of evidence.8

Both strategies for coping with fairness-based exclusions, together with their associated conceptual taxonomies, involve distancing and silencing. Judicial ‘discretion’ is presumptively inferior to ‘proper rules’ in mainstream evidentiary analysis, and frequently attracts direct criticism.9 If you believe that discretionary decision-making is euphemistic code for rulings by instinct and whim, more sophisticated jurisprudential reconstruction is pre-empted. Meanwhile, evidentiary rules characterised as ‘extrinsic’ are by express conceptual fiat banished from the core of legal analysis; at best a sideshow to the main event.10 Predominantly epistemic conceptions of the Law of Evidence also tend to reinforce the awkward and often misleading conventional dichotomy in English criminal jurisprudence between ‘evidence’ and ‘procedure’.11

This essay challenges these widespread disciplinary assumptions, extending an argument expounded by Adrian Zuckerman several decades ago. Zuckerman proposed ‘an explanation of the existing law in terms of [judicial] discretion’ and general principles.12 Once we recognise evidence law for what it really is, Zuckerman insisted, we can grapple with the essential job of articulating normative standards to guide the exercise of judicial decision-making, mindful that ‘the judicial task is a demanding one and that the trial judge is often called upon to make difficult assessments of competing considerations’.13 Terminological confusions may be partly responsible for a certain amount of unfocused initial hostility towards Zuckerman’s thesis. As English courts observe from time to time,14 what we are really concerned with is wise and well-informed judicial judgement rather than free-floating ‘discretion’.15

This chapter develops the thrust of Zuckerman’s insight along three dimensions, using detailed illustrations of evidence obtained by coercion or deception. First, it highlights the importance of moral reasoning in the development and application of evidentiary principles, and—picking up on Zuckerman’s remark about the demands of judging—illustrates some of the complexities16 involved in translating moral reasoning at-large into its applied institutional derivative, criminal jurisprudence.17 This is not to deny the centrality of epistemic considerations in criminal adjudication, but rather to put them in their rightful place: they are neither the exclusive concern of criminal procedure law nor its overriding preoccupation.18

Secondly, this chapter takes up the book’s central theme by exploring the growing impact of human rights norms on common law evidence, a development that was barely conceivable 20 years ago but which turns out to be highly congruent with Zuckerman’s argument. The ‘human rights revolution’ has catalysed a strand of British Evidence scholarship, originally also pioneered by Andrew Ashworth19 and Ian Dennis,20 which places greater emphasis on normative moral reasoning, utilising—much debated—intermediary concepts such as ‘judicial integrity’ and the ‘legitimacy of the verdict’.21 What began as a subsidiary and somewhat submerged theme in British Evidence scholarship is now rapidly gaining recognition and new adherents, contributing powerful restatements of distinctive normative visions.22

Finally, a third dimension of my analysis is comparative and rather more tentative. The following pages contain further examples of migrating norms, transnational judicial ‘conversations’, and institutional pluralism and diversification (driven in particular by the expanding influence of the European Court of Human Rights), intimating another major shift in disciplinary thinking.23 The Law of Evidence as common lawyers have traditionally conceived it is gradually evolving into an increasingly cosmopolitan jurisprudence of criminal procedure, in which overt moral reasoning and a diverse multiplicity of normative sources will become defining characteristics.


Deception and coercion both relate to the criminal trial’s epistemic aspirations in complex ways. People can be tricked into revealing the truth or caught out in a probative lie, such as a false alibi.24 Deception is sometimes an effective investigative strategy for truth-finding. On other occasions, however, it may undermine trust in the veracity of a confession if it was procured through deception or coercion. This elementary point is most obvious in relation to confessions obtained by physical violence or extreme psychological pressure. Everyone is familiar with the idea that people can be tortured or frightened into making a false confession, and psychologists have documented the phenomenon.25 Solzhenitsyn observes that prisoners’ wills can be broken, and rapidly, through the use of entirely prosaic techniques such as sleep deprivation or being made to sit in a chair for days on end.26 These truisms are well-entrenched in English law as a basis for excluding presumptively unreliable admissions and confessions.27

The epistemic connection between coercion and truthful revelations cannot be discounted, however. Many police forces and security services around the world are thought to employ torture routinely28 and for a variety of purposes, some of which certainly include procuring reliable information. Judicial torture has been employed as an epistemic instrument in the past,29 and the intelligence-gathering potential of coercive interrogation continues to be advocated on a cost/benefit analysis by certain governments and commentators.30 There are conceivably limiting cases in which coercion—over and above that already implicit in custodial police detention—does not imperil truthfulness.31 But there are always contingent complications. At some level, the intensity of pain or its fearful anticipation will outweigh competing motivations, including moral or religious scruple, for most people. Victims of extreme torture describe the experience in terms of complete loss of autonomy, personality and self.32 A person can presumably be terrorised into saying just about anything, like Orwell’s Winston Smith in Room 101.

The epistemic implications of deception are even denser and more contorted. Deception might be a good way of tricking offenders into giving themselves away and procuring reliable evidence of their guilt. In the classic police sting or eavesdropping operation, for example, the offender is blissfully ignorant of official involvement and neglects additional precautions to cover their tracks. The sting works best just insofar as the coercive pressures liable to undermine the epistemic integrity of custodial confessions are absent. English case-law offers memorable illustrations.33 However, police deception might equally imperil the reliability of evidence.

Around the world, police officers resort to the same tried-and-tested ruses, unless required to abandon them by robust legal regulation. Suspects are told that it will be better for them to confess early and get it off their chests (a tactic also noted by Solzhenitsyn), which is a snare for those whose overwhelming immediate desire is to be released from police custody.34 Incriminating evidence is manufactured, or its strength greatly exaggerated, extracting false confessions from fatalistic suspects.35 In this way, deception may have epistemic consequences similar to physical and mental coercion—producing false positives as well as true positives. Other deceptive investigative tactics, such as encouraging rape complainants to ‘text’ their alleged assailants in hopes of eliciting admissions or incriminating guilty knowledge in reply,36 may have somewhat unpredictable epistemic implications.37 Outright police ‘entrapment’ raises further, more substantive issues. The burning question here is why the accused did what he did, not—or not only—whether there is reliable evidence to prove it. The legitimacy of entrapment, in other words, concerns the moral limits of the criminal law and the political morality of policing; though, as we explore later in the chapter, it may also be conceptualised doctrinally as a consideration bearing on the fairness of trials.

It is tempting to imagine that the accused is either guilty or innocent, and knows it, so that the only real issue in criminal adjudication boils down to whether the accused decides to tell the truth, or not. But even if we focus reductively on exclusively epistemic considerations, the reality of criminal evidence and proof is far more complex, and precarious, than popular wisdom can conceive. Evidence is a constructive product and active achievement of criminal investigations and prosecutions.38 Any information that a suspect divulges—or is said to have divulged—at any stage of the process will be interpreted in the light of investigators’ stock of other knowledge and objectives and fitted into an appealing theory of the case. To the extent that coercion or deception is involved in these investigative processes, such tactics inevitably increase the ubiquitous institutional risks that information extracted from a suspect will be false, incomplete or prone to interpretative distortion.


Torture involves the most extreme variants of ‘coercive’ investigative measures. Whereas discussions in the past have featured hypothetical terrorists and ‘ticking bomb’ scenarios, post 9/11, real terrorists and exploded bombs have brought the evidentiary status of torture evidence into English courts.

A v Home Secretary (No 2)39 involved conjoined appeals by 10 individuals who had been certified as threats to national security by the Home Secretary and detained without charge pursuant to (since repealed) provisions of the Anti-Terrorism, Crime and Security Act 2001. A person so certified could appeal to ‘SIAC’—the Special Immigration Appeals Commission.40 Lord Bingham formulated the precise question falling for determination by the appellate courts: ‘[M]ay the Special Immigration Appeals Commission… receive evidence which has or may have been procured by torture inflicted, in order to obtain evidence, by officials of a foreign state without the complicity of the British authorities?’41

SIAC itself was prepared to hear such evidence, and this inclusionary approach was upheld by a 2-1 majority of the Court of Appeal.42 But the House of Lords unanimously ruled that SIAC, as a court of law, could not receive foreign torture evidence, even if the Home Secretary was justified in taking such information into account in making the initial decision to certify a particular individual as a terrorist threat to national security. The Law Lords then split on the further issue of the burden and standard of proof. By four votes to three, the majority43 held that alleged foreign torture evidence could be received by SIAC unless it had been established, on the balance of probabilities, that the information was tainted by torture. The minority, comprising the three most senior Law Lords, Lord Bingham, Lord Nicholls and Lord Hoffmann, insisted that SIAC should reject information alleged to be the fruits of torture unless the suspicion of torture could be rebutted on the balance of probabilities.44

(a) Rationalising Exclusion, Beyond Reliability

Arguments for excluding evidence obtained by torture grounded exclusively in considerations of reliability are instrumental and epistemic. Competing arguments appeal to intrinsic moral values, either as comprehensive alternative rationalisations for excluding torture evidence or as side-constraints on epistemic objectives. Appeals to intrinsic values are not self-evidently compelling—a point lost on some of their proponents. Critics might challenge the salience or weight of the values said to mandate exclusion, in light of their deeper philosophical commitments.45 A more common reaction is that torture, though contrary to binding standards of humanity and decency, may still be the lesser evil all things considered, especially if the alternative might imply leaving dangerous individuals at large and unpunished for their crimes. Such arguments do not necessarily collapse into consequentialism, as is often mistakenly believed. Indeed, the possibility of genuine conflicts of values, rights or duties is categorically denied by classical Utilitarianism. Value-conflict is perforce a problem for deontologists.46

Articulated rationales for evidentiary exclusion often blend (and sometimes confuse) instrumental and intrinsic considerations, but it always promotes analytical clarity to try to distinguish between them and to assess their respective cogency and implications. In the context of criminal adjudication, clarity in articulating exclusionary rationales is a vital ingredient of transparency in first instance rulings, and essential for appellate tribunals laying down general guidance for trial judges to apply in future cases. Considerations of reliability, deterrence and intrinsic moral values were all canvassed in their Lordships’ speeches in A v Home Secretary (No 2) as potential rationales for evidentiary exclusion.

The argument that torture evidence should be excluded because it is unreliable was propounded by Lord Carswell:

The unreliability of such evidence is notorious: in most cases one cannot tell whether correct information has been wrung out of the victim of torture—which undoubtedly occurred distressingly often in Gestapo interrogations in occupied territories in the Second World War—or whether, as is frequently suspected, the victim has told the torturers what they want to hear in the hope of relieving his suffering. Reliable testimony of the latter comes from Senator John McCain of Arizona, who when tortured in Vietnam to provide the names of the members of his flight squadron, listed to his interrogators the offensive line of the Green Bay Packers football team, in his own words, ‘knowing that providing them false information was sufficient to suspend the abuse’: Newsweek, 2 November 2005.47

The contrary position, inferring a different lesson from mid-century experiences, was put by Lord Rodger:

Information obtained by torture may be unreliable. But all too often it will be reliable and of value to the torturer and his masters. That is why torturers ply their trade. Sadly, the Gestapo rolled up resistance networks and wiped out their members on the basis of information extracted under torture. Hence operatives sent to occupied countries were given suicide pills to prevent them from succumbing to torture and revealing valuable information about their mission and their contacts.48

In the event, the empirical question of the reliability of information procured by torture did not need to be settled, because none of their Lordships regarded reliability as the controlling factor. As Lord Rodger continued, ‘the torturer is abhorred as a hostis humani generis not because the information he produces may be unreliable but because of the barbaric means he uses to extract it’.49

The ‘deterrence’ rationale for evidentiary exclusion, which is influential in modern US criminal jurisprudence,50 is instrumental and future-orientated, but predominantly non-epistemic. It posits that improperly obtained evidence must be rejected by the courts, irrespective of its reliability, in order to disincentivise officials from further law-breaking in future cases. A blanket policy of inadmissibility supposedly sends an unequivocal message to law enforcement officers that violating or ‘bending’ the rules to procure evidence is futile, and might even result in the loss of crucial information that could have been obtained through lawful means. Regrettably, some criminals must be allowed to go free when the constable misbehaves or ‘blunders’,51 because the deterrent effect of evidentiary exclusion would be jeopardised if the courts started making ad hoc exceptions in individual cases.

Whatever the (doubtful) merits of deterrence as a general justification for excluding improperly obtained evidence, the scenario confronting the House of Lords in A exposes some of its limitations. As Lord Hoffmann observed, the English judiciary, ‘cannot aspire to discipline the agents of foreign governments. Their torturers would probably accept with indifference the possibility that the work of their hands might be rejected by an English court’.52 Lord Rodger added, somewhat testily:

[I]t is no part of the function of British courts to attempt to discipline officials of a friendly country. Besides anything else, the idea that foreign torturers would pause for a moment because of a decision by SIAC to reject a statement which they had extracted verges on the absurd.53

Lord Bingham was not quite so dismissive of instrumental rationales. He thought it ‘very likely that the unreliability of a statement or confession procured by torture and a desire to discourage torture by devaluing its product’ had influenced the international community to include Article 15, stipulating an exclusionary rule, into the UN Convention Against Torture (UNCAT). However, all seven Law Lords delivering speeches in A unequivocally invested intrinsic moral values with primary significance. In the words of Lord Bingham, torture evidence is ‘offensive to ordinary standards of humanity and decency and incompatible with the principles which should animate a tribunal seeking to administer justice’.54

The intrinsic moral argument for exclusion begins with the opprobrium attaching to the practice of torture itself. Their Lordships did not mince their words. Lord Nicholls simply stated, ‘torture is not acceptable. This is a bedrock moral principle in this country’.55 Lord Hope branded torture ‘one of the most evil practices known to man’.56 Taking his cue from Blackstone, Lord Hoffmann spoke of dishonour, corruption, degradation and the censure of enlightened opinion:

The use of torture is dishonourable. It corrupts and degrades the state which uses it and the legal system which accepts it. When judicial torture was routine all over Europe, its rejection by the common law was a source of national pride and the admiration of enlightened foreign writers such as Voltaire and Beccaria.57

Lord Brown called torture ‘an unqualified evil. It can never be justified. Rather it must always be punished’.58

Once it is appreciated that torture is the ultimate jurisprudential pariah, universally condemned on imperative moral grounds, judicial exclusion of evidence tainted by torture follows almost by necessary implication. The judicial process cannot allow itself to be ‘dishonoured’, ‘demeaned’, ‘corrupted’ etc, through any association with a practice as morally decrepit as official torture. Lord Hope found institutional support for this conclusion in a passage in Hume’s Commentaries condemning the barbarity of torture.59 Lord Bingham thought it embedded in English common law, adding that the categorical rejection of evidence tainted by torture ‘is more aptly categorised as a constitutional principle than as a rule of evidence’.60 Lord Bingham then embarked on an extensive review of relevant international law sources, linking Article 15 of UNCAT to ‘the wider principle expressed in Article 69(7) of the Rome Statute of the International Criminal Court’, which states that evidence obtained in breach of the ICC Statute or ‘internationally recognised human rights’ must be excluded if ‘the admission of the evidence would be antithetical to and would seriously damage the integrity of the proceedings’.61 Finally, Lord Carswell summarised the conclusion that all of their Lordships, with the possible exception of Lord Rodger,62 shared:

[T]he duty not to countenance the use of torture by admission of evidence so obtained in judicial proceedings must be regarded as paramount and that to allow its admission would shock the conscience, abuse or degrade the proceedings and involve the state in moral defilement… In following this [exclusionary] course our state will… retain the moral high ground which an open democratic society enjoys.63

(b) The Institutional Morality of Ticking Bombs

None of their Lordships’ speeches in A ventures beyond the ‘bedrock moral principle’ that torture is abhorrent, to trace the deeper normative significance of procedural law as an expression of human dignity and democratic political values.64 Several of their Lordships did, however, seek to explain why reliance on torture evidence is especially inappropriate in judicial proceedings, as opposed to executive decision-making. Bringing the ‘ticking bomb’ scenario out of the classroom and into the courtroom, Lord Brown revisited the moral dilemma posed by information obtained by torture which ‘may on occasion yield up information capable of saving… perhaps many lives’: 65

Unswerving logic might suggest that no use whatever should be made of it… the ticking bomb must be allowed to tick on. But there are powerful countervailing arguments too: torture cannot be undone and the greater public good thus lies in making some use at least of the information obtained, whether to avert public danger or to bring the guilty to justice.

The question for decision in these appeals, Lord Rodger reminded counsel, was not whether the Home Secretary acted improperly in seeking to adduce the contested evidence before SIAC, but rather whether SIAC should have agreed to receive in evidence what the Home Secretary legitimately took into account in certifying the applicants as suspected terrorists. It had been ‘perfectly proper for him to rely on the statement when issuing his certificate’.66 The courts had no basis, observed Lord Hoffmann, for interfering with executive intelligence-gathering and analysis conducted to safeguard national security:

Provided that he acts lawfully, [the Home Secretary] may read whatever he likes. In his dealings with foreign governments, the type of information that he is willing to receive and the questions that he asks or refrains from asking are his own affair.67

The key to this distinction was pinpointed by Lord Brown: ‘the functions and responsibilities of the executive and the judiciary are entirely different’.68 The government of the day must have such powers as it needs to keep its citizens safe, within the limits of a democratic constitution. It follows, said Lord Brown, ‘that the executive may make use of all information it acquires: both coerced statements and whatever fruits they are found to bear’.69 Indeed, the government ‘would be failing in its duty if it ignores whatever it may learn or fails to follow it up’.70 But the judicial role is quite different: ‘Generally speaking the court will shut its face against the admission in evidence of any coerced statement’.71

Lord Hoffmann was inclined to believe that ticking bomb dilemmas are ‘less common in practice than in seminars on moral philosophy’,72 but Lord Nicholls thought it worthwhile to restate the classic moral poser:73

The context is cross-border terrorism. Countering international terrorism calls for a flow of information between the security services of many countries. Fragments of information, acquired from various sources, can be pieced together to form a valuable picture, enabling governments of threatened countries to take preventative steps. What should the security services and the police and other executive agencies of this country do if they know or suspect information received by them from overseas is the product of torture? Should they discard this information as ‘tainted’, and decline to use it lest its use by them be regarded as condoning the horrific means by which the information was obtained? The intuitive response to these questions is that if use of such information might save lives it would be absurd to reject it. If the police were to learn of the whereabouts of a ticking bomb it would be ludicrous for them to disregard this information if it had been procured by torture. No one suggests the police should act in this way. Similarly, if tainted information points a finger of suspicion at a particular individual: depending on the circumstances, this information is a matter the police may properly take into account when considering, for example, whether to make an arrest.74

Lord Nicholls was prepared to concede that any use of torture-derived information by state officials regrettably lends an air of legitimacy to the practice of torture and diminishes its universal prohibition:

[T]he executive arm of the state is open to the charge that it is condoning the use of torture. So, in a sense, it is. The government is using information obtained by torture. But in cases such as these the government cannot be expected to close its eyes to this information at the price of endangering the lives of its own citizens. Moral repugnance to torture does not require this.75

Executive reliance on torture evidence to defuse the ticking bomb is sometimes justified, on this account, as the lesser of two evils, not because such uses are morally unproblematic or free from deleterious side-effects. However, Lord Nicholls went on to insist it would be,

an altogether different matter for the judicial arm of the state to admit such information as evidence when adjudicating definitively upon the guilt or innocence of a person charged with a criminal offence. In the latter case repugnance to torture demands that proof of facts should be found in more acceptable sources than information extracted by torture.76

This nuanced analysis should not be confused with the simplistic ‘separation thesis’ criticised by Andrew Ashworth.77 Differentiation, not separation, of institutional duties, roles and values is the motor driving Lord Nicholls’ train of argument to its ultimate destination: ‘torture is torture whoever does it, judicial proceedings are judicial proceedings, whatever their purpose—the former can never be admissible in the latter’.78


Police ‘entrapment’ is a useful, if conceptually protean, topic for exploring the implications of deception for the admissibility of evidence in criminal trials.79 Since the Human Rights Act 1998 fully entered into force in England and Wales on 2 October 2000, an important dimension of English courts’ approach to allegations of entrapment is the need to ensure that applications of PACE 1984, section 78,80 and common law principles of exclusion and abuse of process satisfy the fair trial standards of ECHR Article 6.81 The following discussion focuses on relevant jurisprudence of the European Court of Human Rights, partly because it is relatively unfamiliar to common lawyers.

On the face of it, the specific fair trial rights enumerated by Article 6 are focused on the trial stage of criminal proceedings. There is no explicit reference to criminal investigations in general, or to police entrapment (or related ideas like undercover operations or covert surveillance) in particular. However, the Strasbourg Court adopts an avowedly flexible and non-formalistic approach to textual interpretation. The Convention is treated as a ‘living instrument’ that must move with the times and adapt to meet the challenges of securing practical rights protection across the 47 Council of Europe states. The ECtHR is not hidebound by its own past pronouncements, especially if, on later reflection, they appear dated or have been overtaken by events.82 The prevailing assumption is diametrically opposite to what the common lawyer would expect: authorities (decidedly not formal precedents) tend to become less pertinent, not more venerable, with the passage of time. When presented with an obliging set of facts, the ECtHR found little difficulty in extending the scope of Article 6 to protect applicants against ‘entrapment’ by police investigators.

(a) Evidence of Entrapment in Strasbourg

In Teixeira de Castro83 Portuguese undercover police officers invited VS, a known hashish user and small-time dealer, to put them in touch with somebody who could supply heroin. After much inconclusive negotiation, VS introduced the officers to the accused, who duly procured heroin from an acquaintance and sold it on to the officers. The accused’s conviction and six-year sentence for drug-dealing were upheld by the Portuguese Supreme Court, which took the view that sacrifices of individual freedom are sometimes justified in order to pursue legitimate law enforcement objectives like tackling the scourge of illegal drug distribution and its associated misery.

The Strasbourg Court, however, upheld the accused’s complaint that he had been denied a fair trial in contravention of Article 6, characterising the undercover officers’ conduct as impermissible ‘incitement’ tantamount to official crime-creation. The operation was further compromised, in the ECtHR’s eyes, because it was not properly authorised or supervised by a magistrate. The Court concluded:

[T]he two police officers did not confine themselves to investigating Mr Teixeira de Castro’s criminal activity in an essentially passive manner, but exercised an influence such as to incite the commission of the offence… [Their] actions went beyond those of undercover agents because they instigated the offence and there is nothing to suggest that without their intervention it would have been committed. That intervention and its use in the impugned criminal proceedings meant that, right from the outset, the applicant was definitively deprived of a fair trial.84

In spite of its technical defects and omissions,85 Teixeira