Hamer, David --- "'Hoist with his own petard'? Guilty lies and ironic inferences in criminal proof" [2001] UQLRS 2; (2001) 54 Current Legal Problems 377-413

Last Updated: 7 October 2009

THE UNIVERSITY OF QUEENSLAND
LEGAL RESEARCH SERIES

‘Hoist with his own petard’? Guilty lies and ironic inferences in criminal proof

(Published in Michael Freeman (ed) (2001) 54 Current Legal Problems 377-413)

David Hamer[*]

‘Hoist with his own petard’? Guilty lies and ironic inferences in criminal proof

David Hamer

The prosecution turns, as it nowadays so often does, to the theory that the accused manifested consciousness of her guilt by telling lies about what happened. In practical terms, what this seems to mean in a case like this is that, having examined the accused’s statements once in search of damaging admissions, it is necessary to examine them a second time with the idea in mind that any exculpatory matter they contain may really be evidence not of her innocence but of carefully concealed guilt.

− Justice McPherson of the Queensland Court of Criminal Appeal[1]

Introduction

The criminal trial is adversarial. The prosecution necessarily advances a different position to the defendant. The defendant may for example say, ‘I was at home with my wife at the time the robbery was committed.’[2] At a minimum, the prosecution will argue that this statement is a lie. The prosecution may go further, claiming that the lie discredits other of the defendant’s statements – the ‘credibility lie’ argument. Or the prosecution may go further still and argue that the lie was told in order to conceal the defendant’s guilt. The defendant’s ‘guilty lie’ is proof of their guilt. This article is concerned with the logical limits of the guilty-lie inference.

The early High Court of Australia decision, Eade,[3] provides a good illustration of the guilty-lie inference. The defendant had been convicted of indecently assaulting a five-year-old girl.[4] The prosecution primarily relied on complainant’s testimony, unsworn on account of her age. She told the Court that, at the defendant’s request, she had bought two pies for him.[5] When she delivered the pies to the defendant he pulled her into the house, pushed her on to the couch, and indecently assaulted her. According the law of the time, the defendant could not be convicted on the basis of the complainant’s unsworn testimony unless it was corroborated.[6] Ironically, the defendant was the source of the necessary corroboration. When confronted with the child, the defendant had denied all knowledge of her visit to his house with the pies. However, pies purchased by the child were found in the defendant’s home, strongly suggesting that the visit had taken place.[7] The High Court held that it was open to the jury to find that the defendant had lied about the child’s visit because he was ‘unable to account for [it] in any way consistent with his innocence’.[8] Had the defendant admitted the child’s visit he would have been immune from prosecution. He was ‘hoist with his own petard’. [9]

Over the last decade, the criminal appeal courts in England, Victoria and other common law jurisdictions have faced an avalanche of appeals concerning the guilty-lie inference. That the inference is the source of many puzzles should not be surprising. The inference has a complex ironical structure: the defendant’s exculpatory statement is given inculpatory effect. In this article I draw upon ironical concepts to explore the conditions governing its valid operation.

In the first part of the article, I break the guilty-lie inference down into its several logical steps, revealing its complexity and fragility. However, these aspects appear not to be appreciated by juries. On the contrary, many authorities comment on the need to direct carefully the jury as to the possible pitfalls of the inference. In Part 2 of the article I suggest that this may be due to the allure of irony.[10] By drawing the inference, the juror appreciates an irony, and enjoys the associated feelings of superiority and cleverness,.

As in all irony, the inference involves a ‘confrontation or juxtaposition of incompatibles’.[11] Where an irony is clear, the literal interpretation can readily be rejected in favour of its unspoken opposite. However, such resolution is not always available. Some ironies ‘take the form of paradoxes, dilemmas, or what we call “impossible situations”’.[12] In Part 3 I seek to lay down some guidelines to deal with the logical doubts that have recently been raised with the guilty-lie inference.

1. Complexity and fragility

It is regrettable that a mystique seems to be developing in New Zealand about lies in criminal cases. The matter is not really complicated.

− Justice Cooke of the New Zealand Court of Appeal[13]

In recent years the appeal courts in England and other jurisdictions have bemoaned the ‘epidemic’[14] of appeals concerning the guilty-lie inference. In response to this, in Victoria, prosecutors have developed a ‘reluctance ... to treat material lies as evidence of guilt.’[15] Ironically this development may benefit both sides.[16] It is to the defendant’s advantage, since ‘it deprives the prosecution of a potent weapon in its armoury’,[17] but the prosecution also benefits, because ‘the jury is likely to attribute [to the inference] the evidential impact which they think it deserves and, in the absence of adequate direction, the accused is at risk of being disadvantaged’.[18] However, the Victorian Court of Criminal Appeal describes it as ‘an unfortunate trend’.[19]

Courts have expressed frustration with the number of appeals generated by the inference, pleading its ‘essential simplicity’.[20] However appeal courts should take some of the blame. In Chan[21] Brooking JA made ironic reference to the Court’s earlier decision in Renzella:[22]

[T]he Court, while deploring that a mystique seemed to be developing about lies in criminal cases, nonetheless found it necessary to proceed to lay down what have in later cases been called twelve propositions but are in fact twelve paragraphs containing many more propositions than that.[23]

But declarations as to the simplicity of the inference verge on the disingenuous. While Brooking JA has suggested that the inference is a ‘very old and simple one, founded on common sense’,[24] the earlier authorities are in fact difficult to reconcile and ambiguous.[25] In 1949, the Privy Council in Tumahole Bereng[26] appeared to be hedging its bets, indicating

[The accused’s] failure to give evidence or the giving of false evidence ... may bear against an accused and assist in his conviction if there is other material sufficient to sustain a verdict against him. But if the other material is insufficient either in its quality or extent they cannot be used as a makeweight.

This seems to place the prosecution in a catch-22. The defendant’s lies may be used to support an inference of guilt only if there is sufficient other evidence to secure a conviction. But if the prosecution had sufficient other evidence, the inference ‘would not be needed at all’.[27]

The guilty-lie inference is a complex one[28] involving, on the most basic analysis, two steps.[29] First, the defendant must be found to have lied, and secondly, the lie must be found to have been motivated by the defendant’s consciousness of guilt. These two steps can be broken down further.[30] The conclusion that the defendant told a lie involves findings that (1) the defendant made the statement; (2) the statement was false; and (3) the falsity was deliberate. Moreover, it is one thing to conclude (4) that the lie had no innocent explanation, and was motivated by a consciousness of guilt. But the further question arises (5) whether the was defendant conscious of guilt of the charged offence or some other misdeed. And then, (6) consciousness of guilt can be distinguished from actual guilt.

The inference is like a chain.[31] If any of these six links is broken, the entire inference fails. In fact, focusing for the moment on the second link, ‘if not met with entire disbelief, [the defendant’s exculpatory statement may be] capable of raising sufficient doubt ... to deprive the Crown of proof’. [32] If there is a reasonable possibility that the defendant’s alibi is genuine then the defendant should be acquitted. This may explain the requirement, accepted by the English Court of Appeal that ‘the lie must be admitted or proved beyond reasonable doubt.’[33]

In Kerim[34] doubt was expressed as to whether the telling of a lie was sufficiently proved. The defendant had been convicted of the rape of his 15-year-old stepdaughter in a hotel room while the two were travelling together. In corroboration of the complainant’s evidence, the prosecution relied on a number of matters, including a guilty lie. The defendant had said the complainant had slept ‘in’ her own single bed that night and not in the double bed with him. But there was prosecution evidence that the single bed appeared not to have been slept in. At trial the defendant explained the appearance of the single bed by suggesting that she had slept ‘on’ the bed, however, the trial judge directed the jury that the defendant’s statement could be used to corroborate the complainant’s evidence. The Court held that the trial judge was in error. ‘To be used as corroboration the statement in question must, among other things, be “clearly” shown to be a lie’.[35] The Court indicated that, in this context, the terms ‘in’ and ‘on’ were interchangable, and consequently, ‘[t]here was no basis upon which a reasonable jury could conclude ... that a clear lie had been told by the appellant’.[36]

Recently in Canada and Australia it has been doubted whether the criminal standard applies to the inference. The High Court of Australia in Edwards[37] held:

Although guilt must ultimately be proved beyond reasonable doubt, an alleged admission constituted by the telling of a lie may be considered together with the other evidence and for that purpose does not have to be proved to any particular standard of proof.[38]

In addition to the requirement of falsity, the defendant’s statement should be found to be deliberately false for the guilty-lie inference to operate. In Ali Ahmed[39] the defendant had been convicted of rape. In a police interview he denied even knowing the complainant, but at trial he admitted intercourse and argued that she had consented on this as on previous occasions. The trial judge warned the jury that ‘[i]t would not be corroboration if it was a lie uttered in your view because of panic, or because of misunderstanding of the question, or for a dozen other reasons.’ The Court of Appeal held that this caution was insufficient. There was strong evidence that the false statement was the result of confusion by the defendant. He was being interviewed through an interpreter and prior to the interview he had instructed his solicitor that he did know the complainant. ‘That factor alone should have alerted this very experienced judge to give a particularly careful direction.’

Most guilty-lie appeals have been concerned, not with the lie, but with the question whether the lie was motivated by the defendant’s guilt of the charged offence.[40] In many cases the court should consider the possibility that the defendant lied to ‘bolster a true defence’.[41] These explanations will have greater plausibility where the lie was told after the defendant had knowledge of being a suspect.

The tendency of persons under suspicion to wish to distance themselves from the persons or events connected with the alleged crimes and to endeavour to improve their position by falsehood is far too common to enable an inference to be drawn with confidence, in any but the rarest of cases, that lies proceed from a consciousness of guilt.[42]

It might be argued that ‘[m]en who are conscious of right have nothing to fear ... [T]he jury will protect them.’[43] But this view ‘has more to do with some idealised vision of universal faith in the perfect administration of justice together with uniform levels of personal fortitude and resolution among those accused of crime, than with reality’.[44] The defendant in Sharp[45] may have lacked such levels of faith and fortitude. He had been charged with fraud in connection with the complainant’s investment of £10,000 in his business. Initially the defendant denied ever receiving a cheque for that amount from the complainant. At trial, however, the defendant admitted receiving the money but denied that it had been obtained by false representations. His counsel’s explanation for the initial lie was that the defendant had been ‘over-egging his defence’.[46]

On other occasions, the defendant may have lied to protect someone else. The defendant in Samuels,[47] on being questioned in connection with drug offences, initially lied about the identity of another suspect and the link between that person and a briefcase full of drugs. The Court acknowledged that these lies could have been a ‘smokescreen ... for the purpose of hiding [the defendant’s] guilt’.[48] However, the lies ‘could just as well’ have been for the purpose of protecting the other person. The lies ‘were certainly relevant to his credibility but they were not capable of adding anything to the Crown evidence against him’.[49]

Even where the defendant’s lies display a consciousness of guilt, this may have no connection with the charges. In Holman[50] the defendant, charged with blackmail and fraud offences, was shown to have given the police a false name. However, as the Court noted, ‘there were before the jury competing explanations for her use of a false name at that time and thereafter’ including her possible status as an illegal immigrant. These lies, therefore, ‘could have no possible bearing upon any issue arising’.

And where the consciousness of guilt has arisen out of the charged offences, this does not always equate with actual guilt. As the Court in Middleton[51] noted,

[A] married man who has had consensual sexual intercourse with a woman and is then faced with an allegation of raping her will sometimes untruthfully deny the act of sexual intercourse at all, in order selfishly to avoid embarrassment to him of his wife’s discovery of his infidelity or, less selfishly perhaps, the consequent anguish that the knowledge may cause to her and their children.

In Dale[52] the defendant had all the more reason to deny intercourse. The lie was told while he was being questioned in the presence of his wife.

It may be inappropriate to describe this kind of explanation for the lie as ‘innocent’. It has clear application to a murder case where the defendant presents the partial defence of provocation. In Richens,[53] Taylor LCJ pointed out:

[A] man who has killed by reason of loss of self-control and therefore faces arrest, trial and possibly lengthy imprisonment, may have almost as strong a reason for attempting to conceal his deed and lie about his involvement as a man who has killed deliberately.[54]

As a generalisation, if the defendant admits some involvement in the events, and the issue is confined to the precise level of involvement or the mens rea, the operation of the guilty inference will be questionable.[55] In Gionfriddo and Faure[56] the Court held that the guilty-lie inference was open even though the defendant’s only defence was directed to intent, however, the Court’s reasoning is puzzling. The defendant had been convicted with his co-defendants of a number of offences in connection with a robbery. He initially denied knowing his co-defendants. At trial he conceded that he did know these men and his involvement in the robbery, but claimed that he had thought it was a fake robbery to support a fraudulent insurance claim. The Court held ‘it is not to the point that at the trial an accused chooses to contest guilt on issues to which an earlier told lie had no, or might not have had any, relevance.’[57] But if the lie had no relevance to the issues at trial, then how could the inference be open?

Above, I considered the requirement that the falsity of the defendant’s statement be clearly established or proved beyond reasonable doubt, in order for the inference to operate. A number of authorities have also applied a higher standard of proof to the latter part of the inference: ‘only if the jury is sure that the defendant did not lie for an innocent reason can a lie support the prosecution case’.[58] As noted above, in Canada and Australia, the highest courts have held that the criminal standard has application only to the ultimate question of guilt.[59] Strictly speaking, if it is certain that the defendant has told a lie, the only requirement should be that the ‘lie is more consistent with guilt than with innocence’.[60] Perhaps, however, a higher standard should be applied as a safeguard, in recognition of the complexity and fragility of the inference. ‘The possibilities of error at all points must be combined and assessed together.’[61]

2. Ironical allure

Though irony risks the reader’s misunderstanding, it flatters understanding as it would the understanding of a juror. Successful irony encourages the reader to join in the inferences and to believe them all the more eagerly.

−Alexander Welsh[62]

As discussed in the previous part, the conditions for the valid operation of the guilty-lie inference constitute ‘a hard test to satisfy’.[63] It is widely believed, however, that juries fail to appreciate the limitations operating on the inference. On the contrary, ‘[t]here is a natural tendency for a jury to think that if an accused is lying it must be because he is guilty and accordingly to convict him without more ado.’[64] The trial judge should recognise that prosecution reliance on lies is ‘fraught with the risk of miscarriage’[65] and direct the jury as to the potential weaknesses in the inference.

The English Court of Appeal had indicated the need for a lies direction wherever the prosecution relies on the guilty-lie inference, and also where there is a risk of the jury invoking it without prosecution invitation.[66] Tucker[67] provides an illustration. The defendant was convicted of inflicting grievous bodily harm in the course of a fight involving two other men. Initially he lied to police about knowing the other parties. The prosecution only relied on the lies as going to credibility. The potential of the guilty-lie inference was weakened by the defendant’s concession that he had been involved in the fight, and his explanation for the lie that he was seeking to avoid getting embroiled in the investigation. The trial judge gave the jury no assistance as to the significance of the defendant’s lie. The defendant’s appeal that a lies direction was required was upheld.[68] Lies featured largely in the case, and the trial judge should have provided ‘a limited direction ... because there is always a risk that a jury may think that because a man has lied, he may be guilty, rather than that lies may simply bear upon his credibility in the witness box.’ As Birch suggests in her commentary, ‘the direction should not depend upon the precise way in which the lies are deployed by the Crown.’[69]

The High Court of Australia appears to have adopted a narrower view. ‘As a general rule, [a guilty-lies] direction should only be given if the prosecution contends that a lie is evidence of guilt ... and if, in fact, the lie in question is capable of bearing that character.’[70] However, the English Court of Appeal is conscious of the volume of appeals on this matter, and in practice the difference in approach may be slight. In Burge and Pegg,[71] Kennedy LJ made reference to Birch’s commentary on Tucker, but then described the inference as ‘a very specific prosecution tactic’,[72] and suggested that no direction was required in the present case since ‘prosecution did not adopt the tactic ... and there was no reason for the judge to think that the jury would themselves approach the evidence ... in that way.’[73] Ideally, the trial judge should ‘consider with counsel whether ... such a direction is in fact required, and, if so, how it should be formulated.’[74] There is also some onus on defence counsel to ‘alert[] the judge ... and ask[] him to consider whether a direction should be given’.[75] If the defendant has not raised the matter at trial, or if the trial judge has proceeded in consultation with counsel, the Court of Appeal will ‘be very slow to interfere with the exercise of the judge’s discretion’.[76]

In Goodway[77] and subsequent cases, the Court of Appeal has recognised a further qualification to the need for a lies direction, suggesting that it is not required where there is a ‘straight conflict of evidence’[78] and the direction would be ‘otiose’[79]. Ironically while this principle appears to be a further attempt to stem the flow of appeals, its extent is unclear, and it may have the opposite effect.[80] Two distinct strands of authority are discussed in Section 3.5 below.

The burgeoning case-law on the guilty-lie inference raises the question, why does the inference present ‘special risks of illogical intuitive reasoning’?[81] Why do jurors fail to appreciate its inherent weaknesses? A similar question arises with respect to evidence of a defendant’s other misconduct, also known as similar fact, tendency or propensity evidence. That evidence of a defendant’s other misconduct may have some probative value is now accepted.[82] However the evidence carries a number of risks. ‘[J]urors would in many cases think that [the evidence] was more relevant than it was’,[83] and ‘the facts of the other misconduct may cause a jury to be biased against the accused’.[84] It is as true of misconduct evidence as it is of the defendant’s lies that such evidence ‘provides potent (and potentially prejudicial) support for the Crown case.’[85]

Just as the jury should be warned, for example, not to misuse use their knowledge of a defendant’s previous conviction, so when the defendant has lied, in order to avoid the prohibited reasoning, the jury will often need to be warned [or] reminded ... of the reality, namely that an innocent defendant may sometimes lie and that the inference of guilt does not automatically follow.[86]

In both cases the jury should be warned not to follow ‘forbidden reasoning’.[87]

In many situations, the inference of consciousness of guilt of the particular crime is so uncertain and ambiguous and the evidence so prejudicial that one is forced to wonder whether the evidence is not directed to punishing the “wicked” generally rather than resolving the issue of guilt of the offense charged.[88]

Sir Edward Coke suggested that even the defendant’s innocent lies may result in conviction through divine intervention. Such miscarriages of justice would provide a caveat ‘to the innocent and true man, that he never seek to excuse himself by false or undue means, lest thereby he offending God (the author of truth) overthrow himself’.[89]

This analogy between the guilty-lie inference and propensity reasoning has some plausibility. However, the misdeeds that are under consideration in similar fact cases – murders, rapes and so on – are of a different order of wickedness than the lies alleged in the guilty-lie cases – the false alibi, for example More is required to fully explain the fatal allure that the guilty-lie inference holds for juries. Perhaps it is the ironical nature of the inference that makes it so appealing.

The ironical nature of the guilty-lie inference is clear. It possesses the three ‘formal requirements’ of irony identified by DC Muecke in his seminal work, The Compass of Irony.[90] First, the inference is a ‘double-layered or two-storey phenomenon’. The defendant, residing at the ‘lower level’, intends their exculpatory statement to be taken at face value with an acquittal resulting. However, there is further evidence or background information[91] that suggests to the juror that ‘the situation’ should not be taken as it is ‘presented at the lower level’. At an ‘upper level’ the juror takes the defendant’s false exculpatory statement as evidence of guilt. The second requirement is the ‘opposition between the two levels’. And thirdly, there is a ‘victim’ of the irony. The defendant was ‘confidently unaware’ that their exculpatory statement could be used to secure their conviction.[92] The defendant is ‘hoist with his own petard’.[93] However, unlike the dramatic irony in Hamlet, the guilty-lie inference is an irony of events, completed through the observation of the ironic juror.[94]

The attraction that the ironic inference holds for jurors can be found in narrative models of juridical proof. The guilt of the defendant may be considered ‘plausible’ because it draws the defendant’s exculpatory statement and the conflicting prosecution evidence into a ‘coherent story’.[95] The danger is, however, that ‘[t]he mind [can be too] apt to take a pleasure in adapting circumstances to one another, and even straining them a little, if need be, to force them to form parts of one connected whole’.[96] Michael Dibdin’s fictional detective, Aurelio Zen, appreciates this risk. ‘The temptation to fit everything into a neat pattern ... should be resisted. It might well be that two or more quite unrelated patterns were at work.’[97] At the end of Umberto Eco’s Name of the Rose,[98] Brother William concedes that he had ‘conceived a false pattern to interpret the moves of the guilty man.’[99]

Among narrative strategies of ‘legitimizing coincidences’[100] irony offers special pleasures: ‘feelings of superiority, freedom and amusement and symbolically as looking down from a position of superior knowledge or power.’[101] These arise out of the ‘contrast between the single and limited vision of the victim and the double, complete vision of the ironic observer’.[102] Sometimes, however, the enthusiastic ironist may see an irony that is not there. They are then ‘equally under judgment, if not by [their] fellow men then by the mocking spirit of irony’.[103] The guilty-lie inference has considerable allure, but the conditions for its correct use are rarely satisfied. All care must be taken to avoid the bitterest of ironies, the ‘searing injustice and consequential social injury ... when the law turns upon itself and convicts an innocent person.’[104]

3. Circularities

Yossarian saw it clearly in all its spinning reasonableness. There was an elliptical precision about its perfect pairs of parts that was graceful and shocking, like good modern art, and at times Yossarian wasn’t quite sure that he saw it at all.

−Joseph Heller[105]

So far in this article it has been assumed that, within certain limits, the inference is a legitimate tool of juridical proof. In this part I consider questions that have been raised as to the rationality of the inference. Is it logically valid for the defendant’s exculpatory statement to be given inculpatory effect? The prosecution tactic has the appearance of trying to head north by walking south.[106]

The phenomenon of irony may give us confidence that, provided the signs are correctly read, the inference is legitimate. It will be appropriate to give the defendant’s exculpatory declarations inculpatory effect. The inference operates with,

corrective or normative irony ... an engineer is hoist with his own petard to the great diversion of the spectators upon whose heads he will not fall ... When the victim is dealt with the incident is closed, the irony is over. In these instances of irony the victim is isolated; he is “in the wrong” and over against him are the rest of society or mankind who are “in the right” and safe.[107]

However, irony does not always lend itself to a preferred interpretation. More modern forms of irony are ‘more self-conscious, more tentative (lacking the element of resolution), and more open to dialectic exposition’. [108] Opposing alternatives are set up but without resolution, leading to ‘some continuous process or closed system, a self-serving mechanism, a phantom state, or indeed an imperspicuously vicious circle.’[109]

In recent years the guilty-lie inference has generated doubts of this kind. An inculpatory inference can be discerned from the defendant’s statement, but so too can the more obvious exculpatory inference, with neither enjoying superiority. These two may be seen as ‘representative’ of larger structures: ‘the conviction that “there is more here than meets the eye” and the suspicion that there is less.’[110] Rather than being a tool of proof, irony may then become vehicle of doubt, not only about the defendant’s guilt in the instant case, but about the very possibility of proof. In the remainder of this part I will consider five different expressions of the circularity to which the guilty-lie inference appears prone.

3.1 Infinite regression

One logical problem that the inference presents is infinite regression. This possibility can be simply demonstrated. The inference takes evidence that appears exculpatory and gives it inculpatory effect. A false alibi, for example, may be taken as proof of guilt. However, an obverse inference can also operate. The defendant’s failure to produce an alibi is proof of their innocence. In Roberts[111] the Court indicated:

[T]he appellant’s conduct after he had been accused was consistent with the honesty of his denial ... in particular he did not deny that he might have been at the dance hall on the evening in question or seek to set up an alibi. He simply said that it was so long ago that he could not remember where he was.[112]

But if the inference can switch in both directions, then it appears fundamentally unstable. Exculpatory evidence can become inculpatory evidence only to become exculpatory once more. When does the flipping mechanism cease? A guilty defendant may have regard to Roberts and not present an alibi in order to avoid suspicion. Once authorities recognise this ploy, the absence of an alibi may be treated as evidence of consciousness of guilt. On learning of this new development a guilty defendant may be tempted to present a false alibi once more. ‘Irony in itself opens up doubts as soon as its possibility enters our heads, and there is no inherent reason for discontinuing the process of doubt at any point short of infinity.’[113]

Another example is provided by Kerim[114] – the case discussed earlier in which the defendant was convicted of the rape of his 15-year-old step-daughter. In corroboration of her evidence, the complainant provided a description of the condoms the defendant used, which, she said, he had bought at the local chemist. She indicated the words ‘for her pleasure’ appeared on the packet, and condoms roughly matching this description were found in the local chemist.[115] (Perhaps the complainant was struck by the irony of juxtaposition.[116]) McPherson J suggested that the slight flaws in her recollection of the condoms lent her story greater credibility: ‘If it was part of a carefully contrived plot to entrap the accused, one might expect her to have memorised the whole of the inscription “textured for her added pleasure” and not simply “for her pleasure”.’[117] One is tempted to add a further reversal. If the victim had carefully contrived to entrap the accused, she may have deliberately introduced the flaws to provide added realism.[118]

In most cases, infinite regression will be more a theoretical than a real risk. In practice it will appear quite clear where the regression should be brought to a halt.[119]

3.2 Corroboration circularity

A specific and well recognised risk of circularity is in connection with guilty lies as corroboration.

If a witness required to be corroborated is believed in preference to the accused and this alone establishes the lie on the part of the accused, reliance upon the lie for corroboration would amount to the witness corroborating himself. That is a contradiction in terms.[120]

For this reason, the Court of Appeal in Lucas (Ruth)[121] laid down the requirement that ‘the statement must be clearly shown to be a lie by evidence other than that of the [witness] who is to be corroborated’.[122] If this requirement is followed, then the corroboration circularity will be avoided.

3.3 The presumption of guilt?

A more general concern about circularity is sometimes expressed about the guilty-lie inference. In Edwards,[123] Brennan J considered the claim that the guilty-lie inference possessed an ‘inherent circularity’.

If the jury must be satisfied that the motive for the lie is a realization of guilt and a fear of the truth, the jury must be satisfied that there is guilt to be realized before it decides whether to treat the supposed lie as corroboration of other inculpatory evidence. If the jury is satisfied of the accused’s guilt, it would be unnecessary to consider whether his evidence is corroborative of evidence tending to prove his guilt.[124]

Martin suggests that the inference ‘is bootstrapped on a presumption of guilt. This error occurs because the reasoning is circular – that the accused is probably guilty is confirmed by his guilty behaviour.’[125] I do not find these explanations of the circularity persuasive. There appear to be two related concerns, linguistic and logical, both of which can be resolved.

One concern is with the traditional term – ‘consciousness of guilt’. This expression ‘labels the conduct in question in a preconceived manner’.[126] It ‘suggest[s] a conclusion about the conduct in question which undermines the presumption of innocence and may prejudice the accused in the eyes of the jury.’[127] This criticism may be warranted, but it is directed to the label rather than the rationality or logic of the inference, and is easily overcome. It should be recognised that the application of the label ‘“consciousness of guilt” means no more than that the conduct in question provides some evidence of guilt.’[128] If felt necessary, the label could be changed to ‘evidence of post-offence conduct’.[129]

There is, however, a stronger logical version of this objection, which is associated with the claim that the criminal standard should be applied to inference.[130] Since the inference presumes guilt, and guilt must be proved beyond reasonable doubt, then the defendant’s consciousness of guilt should also be proven to that standard for the inference to operate. This argument appears to motivated by ‘the direction of causality’.[131] The defendant’s commission of the offence gave rise to their consciousness of guilt, which gave rise to their lie. There is a tendency for human inference to follow this chronology, and so the finding that the defendant is conscious of their guilt seems to presume their actual guilt. It appears that ‘the jury must conclude that the accused was guilty before it can use the telling of a lie by him as evidence against him’.[132] Consider how much more natural it is to draw an inference from motive evidence, where the steps of the inference follow causality. The defendant’s motive may have caused them to commit the crime. Evidence of motive therefore increases the probability that the defendant committed the crime. There is no appearance of a presumption of guilt.

The direction of causality provides an explanation for the presumption-of-guilt objection, but it does not provide validation. Inference can operate logically against the direction of causality. In an appropriate case it can be accepted that a guilty person is more likely to lie than an innocent person. It logically follows that evidence that a defendant has lied increases the probability of the defendant’s guilt.[133]

3.4 The simple denial

In the sections above, I conceded that the guilty-lie inference poses the risk of corroboration circularity, however, I rejected the view that there is a general risk of circularity. In this section I will consider a second clear case of circularity – where the alleged lie is a simple denial. Courts have recognised that by no torturing of the statement ‘I am innocent’ can you extract the statement ‘I am guilty’.[134] A finding of guilt would be required to prove the defendant’s lie. And it would then be circular for the finding of a lie to provide further support to the ultimate finding of guilt. The prosecution would be pulling itself up by its bootstraps.[135]

If the guilty-lie inference were allowed from a simple denial, the defendant would be placed in a catch-22, an irony of dilemma.[136] Clearly a confession would be employed by the prosecution to secure a conviction. But so too would a denial. It would be inappropriate in such a case to say the defendant was ‘hoist with his own petard’. The true victim of the ‘hoist’ irony is being punished as a consequence of the choices they have made. The defendant ‘would have been safe had he been less hubristic or more circumspect’.[137] The victim of the catch-22 irony, however, has ‘no exit ... no escape ... [I]t would seem of little matter what anyone tried to do, as the irrationality of the principle of catch-22 would mean that no choice is meaningful and free will is a delusion.’[138] The inference would not be normative or corrective.[139] On the contrary, the inference would be an abuse of power: ‘justice is mocked, the innocent are victimised’.[140]

It is difficult to find actual cases where this illogical unjust operation of the guilty-lie inference on simple denials has been judicially endorsed,[141] although the related ‘central issue’ principle has recently generated a considerable body of case-law, discussed in the next section. In the remainder of this section I will consider examples from crime literature and a famous Texan miscarriage of justice, which illustrate the corrupt nature of the inference.

In Umberto Eco’s Name of the Rose, Bernard Gui conducts an inquisition of Remigio to determine whether he is a Pseudo Apostle, a heretic. Remigio swears that he, like Gui, is a follower of the ‘Roman church, one, holy, and apostolic, governed by the Pope and his bishops’.[142] Gui takes this denial as a sign of guilt, telling Remigio, ‘I know very well that you Pseudo Apostles grant dispensations to those who swear false oaths rather than betray the sect. And so every oath will be further proof of your guilt!’ Gui is quite open about the dilemma he imposes on Remigio. ‘And you will be damned and condemned if you confess, and damned and condemned if you do not confess, because you will be punished as a perjurer!’[143] As Parker notes, Gui’s ‘search for truth is a pragmatic and expedient method of furthering the aims of the papal court at Avignon. ... His conclusions are set; only the means of execution are in question.’[144] Colletti, in an unrelated book of the same name published in the same year,[145] makes a similar point, noting that Gui is presented as ‘the “scourge of heretics” who will manipulate meanings and appearances to serve the powerful interests of the pope who he represents.’[146] However, Colletti may be doing Gui an injustice to suggest that ‘the double meanings on which ... irony depends have no place in Gui’s universe.’[147]

Another of the guilty-lie inference operating on a simple denial is found in Michael Dibdin’s Vendetta. The investigation of crime in Dibdin’s late twentieth-century Italy appears not altogether different from that in Eco’s early fourteenth-century Italy. Crimes were occasionally solved ‘but that was incidental to their real purpose, which was to maintain or adjust the balance of power within the organization itself.’[148] Dibdin’s Venetian detective, Aurelio Zen, has a not entirely undeserved reputation as an ‘unscrupulous manipulator of evidence and witnesses’.[149] In, Vendetta, Zen is accosted by Spadola, who claims that he was framed by Zen and unjustly imprisoned for 20 years. He says to Zen,

I told you I was innocent. I told you I hadn’t done it. I knew I’d been betrayed, and that made it all the harder to bear. ... And do you know what you said, when I screamed my innocence in your face? You said, “Yes, well you would say that, wouldn’t you?” And you looked at me in that sly way you educated people have when you’re feeling pleased with yourselves.[150]

Further examples of the guilty-lie inference operating on a simple denial can be drawn from Errol Morris’s 1988 film, The Thin Blue Line.[151] Through extensive interviews with the key participants, Morris tells the story of the investigation, prosecution and conviction of Randall Adams for the murder of a police officer in Dallas, Texas, on 29 November 1976. As one watches the film, it becomes apparent that the true murderer was David Harris, who had appeared as an eyewitness for the prosecution. In a final interview with Harris in December 1986 while he was on death row for a different murder, Morris asked Harris whether he thought Adams was innocent.

DH: Have you asked him?
EM: Well, he’s always said he’s innocent.
DH: There you go. Didn’t believe him, huh? Criminals always lie.

Harris then virtually confesses. He says he is ‘sure’ Adams is innocent ‘cause I’m the one that knows’.

Harris was, in fact, the obvious suspect: he matched a police witness’s original description; the shooting was with a weapon he had taken from his father; it was from a car he had stolen from a neighbour; it took place while he was on a violent crime spree; and soon afterwards he bragged about the killing to acquaintances. But, as Harris later said to Morris, he was a ‘scared 16-year-old kid, would surely like to get out of it if he can’. Harris denied the crime to police, claiming that he had picked up Adams hitchhiking, and that Adams had fired the shot. Adams had no history of violence, but he was a preferable suspect for two reasons. First, as the trial judge noted he was ‘only a drifter’, whereas Harris was a local boy, and the District Attorney was ‘not that keen on ruining a young man’s life’. Secondly, Adams was 28 and a candidate for the death penalty, while Harris at 16 was too young. Since this case concerned the killing of a police officer – in the words of the District Attorney, ‘the thin blue line separating the public from anarchy’ – the authorities were keen for the ultimate sanction.

Asked in his final interview whether he was surprised that the police believed his story, Harris said, ‘I was hoping they’d believe me, you know. After all was said and done, it was kind of unbelievable, you know. But there it is.’ Adams’s denials, however, counted against him, the police considering that ‘whatever he had done, it never really bothered him’, he ‘didn’t have very much conscience’ and that he had ‘overacted his innocence’. Strangely, they never considered the possibility that Adams was reacting genuinely when, as they noted, ‘he protested that he hadn’t done anything. He couldn’t imagine why he’d been brought in.’ When Adams ‘didn’t remember anything about a shooting; he didn’t remember anything about a police officer’, the police considered that ‘that part of his mind just conveniently went blank’. ‘Everything else he remembered vividly and that’s just a convenient memory lapse is all that is.’

The guilty-lie inference subjected Adams to further injustice on sentencing. His attorney clearly felt the absurd irony:

In death penalty cases we have a question ... whether or not [the defendant] has a dangerous mentality and might be expected to commit other crimes. To answer that question the District Attorney’s Office sends in a psychiatrist to the defendant’s cell to discover whether he is without remorse and therefore has a dangerous and psychotic personality. But of course, in the instance of a person that did not commit the crime, naturally they’re not going to show much remorse.

Adams was given the death penalty.[152]

In all of these simple-denial cases the guilty-lie inference presumes guilt, and is circular and invalid. The suspect/defendant is presented with a dilemma – even their declaration of innocence is used as rope to hang them by. By invoking the inference, the authority is abusing their power. The inference appears irrational, unjust and corrupt – and yet not wholly devoid of ironic allure.

3.5 Lies on the central issue

In recent years courts have recognised that the risk of circularity in the guilty-lie inference extends beyond simple denials. The inference is precluded not only where the alleged lie is ‘I am innocent’, but also where it concerns an ‘essential’,[153] ‘central’[154] ‘primary’[155] or ‘main’[156] issue. In such a case ‘the jury cannot decide whether he is lying, without first being satisfied on the central issue of guilt’,[157] in which case, ‘the exercise would be circular, proof of guilt proving the lie, and the lie unnecessarily being advanced as evidence of guilt.’[158] The inference validly operates only on denials concerning ‘collateral’,[159] ‘separate’,[160] ‘peripheral’,[161] ‘secondary’,[162] ‘discrete’,[163] or ‘distinct’[164] issues. Then the prosecution can establish the lie ‘independently of proof of its case’.[165]

The central-issue principle appears clear enough in theory. However, difficulties arise in practice – how is it to be determined whether an alleged lie concerns a central issue so as to preclude the operation of the guilty lie inference? In this section I will first discuss several cases where the central-issue principle had clear application, and then consider a number of cases where its application is far less clear. The latter cases may in fact be applications of a distinct principle with looser application, reflecting a concern with the volume of appeals, rather than the strict logic of the inference.

The strict central-issue principle

A case where the central-issue principle had clear and straightforward operation was Liacopoulous.[166] The appellants were convicted on conspiracy and fraudulent evasion charges in connection with the importation of drugs. According to the prosecution, the appellants ran an English language school as a front for their drug importation. Two of their foreign students were convicted on drug importation offences. In police interview the appellants had admitted assisting the students with their admission and travel plans, but ‘said that they had no knowledge that any of these persons would seek to import cocaine ... and they had nothing to do with drug smuggling.’ The prosecution argued that these denials were false, but it was held that no lies direction was required. This was not a case where the appellants were alleged to have lied ‘on one or more discrete issues’, nor were ‘the lies ... being used as supporting evidence’. The alleged lie concerned the ‘essential question’ in the case: whether the appellants ‘knew all about and were parties to the alleged agreement to import cocaine’. To find that the defendant had lied would amount to a finding that the defendant was guilty, and the situation was ‘covered by the general direction about the burden and standard of proof.’

A further situation where the central-issue principle has a fairly clear application is where the trial boils down to a battle of credit between the complainant or key prosecution witnesses on the one hand and the defendant on the other. In House and Meadows[167] the complainants alleged they had been used as underage prostitutes by the first defendant, and identified the second defendant as their controller. The complainants claimed that the second defendant knew their ages. Indeed their youth was a selling point with the first defendant. The second defendant denied that she knew the first defendant or the ages of the complainants. It was held that no lies direction was required, as the guilty lie inference was not open. The ‘only way’ the defendant could be considered to have lied was ‘by preferring and believing the evidence of the two girls rather than the evidence of the appellant.’ But then ‘the outcome of the case in any event would have become a foregone conclusion’. The jury’s finding of lies was not ‘separate from their determination of the main issue in the case which turned upon the credibility of the complainants’. ‘[T]here was no independently established lie.’

The loose central-issue principle

The application of the central-issue principle beyond simple denials and head-on witness conflicts is far from clear. In principle, following the reasoning above, the determinative question should be whether the lie’s proof turns on proof of guilt. If so, then the guilty-lie inference would be circular and logically invalid. However, courts have applied the central-issue principle more broadly than the foregoing analysis suggests. This may be a sign that a distinct looser central-issue principle is at play.

A useful case in determining the extent of the central-issue principle is Holman,[168] in which the Court applied the principle to certain alleged lies while rejecting its application to other alleged lies. The defendant was convicted on charges of blackmail and forgery in connection with large payments to the defendant from the complainant’s bank account. The complainant was a wealthy person with a shady past including likely tax evasion. The prosecution’s case was that the defendant blackmailed the complainant in connection with his tax difficulties obtaining a payment of £75,000, and then she obtained a further £150,000 by fraud. In her defence the defendant indicated that the first payment was a reimbursement for expenses that she had incurred while helping the complainant with his tax difficulties and that the second payment was a gift made in gratitude for her assistance. ‘The Crown’s case was that her explanation was totally untrue’ but it was held that no lies direction was required. Citing Goodway[169] and Dehar[170] the Court indicated that if the jury rejected the defendant’s explanations for the payments as untrue, ‘their rejection of the explanation almost necessarily would leave the jury with no choice but to convict as a matter of logic’[171] and a lies direction would therefore be ‘otiose’.[172] There were, however, other ‘particular lies within the overall exculpatory explanation proffered by the defendant’ which did call for a more complete lies direction. For example the defendant had claimed to have sought the advice of a specific accountant employed by a particular bank, whereas the prosecution provided evidence that the bank employed no such person.

The Court appears to have drawn the distinction noted at the commencement of this part – the defendant’s innocent explanation for the payments was an alleged lie on a central or essential issue while the particular lies concerned discrete or separate issues.[173] But while the term ‘central issue’ may seem applicable to the defendant’s innocent explanation for the payments, the application of the guilty-lie inference to these statements presents no obvious circularity. In this respect the application of the central-issue principle differs from that discussed in the previous section. The defendant’s statements were not simple denials of certain elements of the offence. The defendant provided a positive innocent explanation for the payments that were made to her. Nor did the case resolve itself into a head-on conflict between the defendant and key prosecution witnesses. On the contrary, the prosecution relied upon a number of witnesses, and even on appeal, the Court found it ‘necessary to recite in a little detail what the evidence revealed to be a rather long and tangled story’.

It seems that a jury in Holman could have rejected the defendant’s explanation for the payments as a deliberate falsity without first accepting the prosecution’s inculpatory account. The conclusion that the defendant’s explanation was a lie could then be used as a foundation for the guilty-lie inference without circularity. Nevertheless, the term ‘central issue’ appears to have application to this looser principle. And, as with the stricter principle, it can be said that ‘the rejection of the explanation given by the accused almost necessarily leaves the jury with no choice but to convict as a matter of logic.’[174] But there is a difference in the reasoning in the two types of case underlying this shared description. Where the strict principle applies the statement that has been found false is ‘merely the obverse of the allegation that the defendant is guilty. In other words, the jury cannot decide whether he is lying, without first being satisfied on the central issue of guilt.’[175] The inference would therefore be circular and invalid. Where the loose principle applies, once the defendant’s story is rejected as untrue ‘then it automatically follow[s] that the appellant must be guilty.’[176] It is not so much that the inference is invalid, but that it is ‘unnecessary’[177] since ‘[t]he very matter ... being said to be a lie was at the very heart of the defence. If the jury were rejecting that matter they were in effect rejecting her defence.’[178] Of course, ‘the mere fact that the jury rejects the defence does not mean that they may convict without more. They still have to be sure that guilt of the specific offence charged has been proved’.[179] It is suggested that ‘the consequence of the jury rejecting the defendant’s story is usually covered by the general directions of law on the burden and standard of proof’. [180]

A recent case which highlights the distinction between the strict and loose versions of the central-issue principle is Middleton.[181] The defendant was convicted of a burglary in which a car was also stolen. The defendant had been caught speeding in the stolen car twelve hours after the burglary, and a subsequent search of his mother’s place where he was residing turned up further stolen property from the burglary. The defendant’s explanation given in evidence was that a man had left the car and the property at his mother’s, and directed him to take the car out of town. The prosecution, necessarily, disputed the truthfulness of this story, but it was held that no lies direction was required. Not, however, on the basis that the guilty-lies inference would be invalid. On the contrary, such an inference was clearly open.

If the jury concluded that the defendant had lied to them when explaining his innocent involvement with the car, his driving of it and the presence of property from the burglary at his mother’s house, the obvious inference was that he was lying to the jury in order to conceal his guilt.[182]

The Court applied the looser version of the central-issue principle. If the defendant was found to have lied, the guilty-lies inference would be unnecessary. ‘[A]s a matter of common sense, in the absence of an innocent explanation to which the jury might give some credence, the Crown’s case that the man in possession of the car was the burglar was fairly formidable’.

Why the distinction matters

This distinction between the loose and strict versions of the central issue principle has not been noted in the English authorities. They all trace their lineage, more or less explicitly, to the Dehar[183] formulation quoted in Goodway.[184] In both sets of cases the principle has been employed to dismiss defendants’ appeals that a lies direction should have been given.

In recent cases the loose principle has been eagerly seized upon as a means of stemming the ‘steady and almost unstoppable stream’[185] of appeals on lies. The defendant’s demands for a detailed lies direction is seen as overly analytical. The summing up is ‘emphatically’[186] not ‘supposed to concern itself with legal theory’,[187] and such a direction would only ‘serve[] to confuse and complicate an essentially simple issue’.[188]

The Court’s dismissive and anti-analytic attitude may endanger the correct application of the strict central-issue principle, which requires close attention to the logical structure of the inference. The risk is particularly grave where the defendant appeals on the basis that the jury may have employed a circular guilty-lies inference. Such an appeal has not yet been heard in England, perhaps because of the failure to distinguish between central issues in the loose and strict senses – between unnecessary and invalid inferences. This problem has, however, been addressed by the Australian courts.

In the New South Wales case, Li Zheng[189] the defendant was convicted on drug supply charges. Police witnesses claimed to have seen the defendant dealing in drugs in a car park, whereas the defendant claimed to have been washing a car and talking with a passer-by. The trial judge directed the jury that if the defendant’s account was rejected it could be inferred that the defendant had a consciousness of guilt. The Court held that the direction was wrong. ‘[T]he only logical way in which the jury could have been satisfied that the appellant was lying was if they accepted as true the evidence of the Crown’s witnesses’,[190] but this would have established the defendant’s guilt.

For the jury to be invited to conclude that the appellant’s lies concerning his conduct in the car park was available as evidencing a consciousness of guilt, which could then in turn be taken into account in some way in determining whether the appellant had in fact conducted himself in that way is a wholly circular argument.[191]

Where the defendant argues that the guilty-lie inference is invalid the distinction between the strict and loose versions of the central-issue principle becomes crucial. There is a great difference between the jury employing an unnecessary inference and the jury employing a circular inference. This is illustrated by the Victorian case Jan Ha Le Trinh.[192] The defendant had been convicted of involvement in heroin importation. She provided an ‘innocent’ explanation for her activities. She had no knowledge of the heroin and had thought she was merely laundering money. The prosecution argued that her story was a guilty lie, and the defendant appealed on the basis that the inference was not open. The defendant’s story was ‘central to her defence’ and the guilty-lie inference would therefore involve ‘a process of circular reasoning’.

The defendant’s arguments were rejected by an application of the strict central-issue principle. The Court acknowledged that a guilty-lie inference could not be founded on a ‘simple denial ... because the lie could only be established by proof by the prosecution of the crime charged.’ This would be a ‘bootstraps argument’. But in the present case, the defendant had advanced the money laundering theory. She ‘went further’, providing ‘an explanation rather than a denial’. The prosecution was correct in their contention that the falsity of the explanation could be found without a finding of guilt. Falsity could be proved with ‘other established evidence in the case’.

There is a crucial distinction between invalid and unnecessary guilty-lie inferences. If this is not appreciated there will be a grave risk that juries will convict on the basis of circular inferences on central-issue lies. This risk is all the more grave given current climate of frustration with the volume of guilty-lie appeals.

Conclusion

In this article I have drawn on the concept of irony to analyse the guilty-lie inference and explore the legal problems it has generated. Where a jury considers that a defendant has lied and no innocent reasons for the lie are apparent, the jury may consider that the defendant was lying to conceal their guilt. This area of evidence law has recently generated numerous appeals in England and other jurisdictions. Unlike the courts, I do not find this surprising. The inference is ironic, which is to say it is fragile, alluring and potentially circular. Where courts have recently urged the ‘essential simplicity’[193] of the inference, my analysis has revealed layered complexity. I have not been able to provide a magic bullet against the innumerable difficulties presented by the guilty-lie inference. I have, however, suggested that an ironical perspective may assist in mapping its deceptively rugged terrain. ‘“If one must warn against irony as a seducer, one must also praise it as a guide” and acknowledge its function as a “disciplinarian”.’[194]

---

[*] BSc, LLB (Hons), Lecturer at the University of New England. Preparation of this article was considerably assisted by my presentation of seminars at University College, London, and the University of Nottingham. For the organization of these I would like to thank Michael Freeman and Paul Roberts respectively. Thanks also to Di Birch, Ian Dennis, Mike Redmayne, Diane Hamer, Geoff Hamer, Ian Hodges and Felicity Plunkett for valuable discussions and suggestions.

[1] Finn (Unreported, Qld CCA, 4 February 1994), 20; quoted by A. Palmer, ‘Guilt and the Consciousness of Guilt: The Use of Lies, Flight and other ‘Guilty Behaviour’ in the Investigation and Prosecution of Crime’ [1997] MelbULawRw 3; (1998) 21 Melbourne University Law Review 95, 121.
[2] Eg, Whitby [1977] 1 QB 224.
[3] [1924] ArgusLawRp 24; (1924) 34 CLR 154.
[4] Crimes Act 1900 (NSW) s 77.
[5] [1924] ArgusLawRp 24; (1924) 34 CLR 154, 158.
[6] Crimes Act 1900 (NSW) s 418.
[7] [1924] ArgusLawRp 24; (1924) 34 CLR 154, 158. This opportunity evidence by itself was held to be insufficient confirmation of the actual commission of the offence: (Knox CJ, Gavan Duffy and Starke JJ), at 159 (Isaacs and Rich JJ). The child’s prompt complaint was also considered insufficient. The child ‘could not be her own corroborator’. Ibid. 157 (Knox CJ, Gavan Duffy and Starke JJ).
[8] Ibid. (Knox CJ, Gavan Duffy and Starke JJ), at 159 (Isaacs and Rich JJ). The trial judge had directed that prompt complaint could provide corroboration, and for that reason a retrial was ordered. Ibid. 158 (Knox CJ, Gavan Duffy and Starke JJ), at 159 (Isaacs and Rich JJ).
[9] There’s letters seal’d: and my two schoolfellows,
Whom I will trust as I will adders fang’d,
They bear the mandate; they must sweep my way,
And marshal me to knavery. Let it work;
For 'tis the sport to have the engineer
Hoist with his own petard: and ‘t shall go hard
But I will delve one yard below their mines,
And blow them at the moon (Hamlet, (III),(iv))
Claudius had killed Hamlet’s father and married his mother thus become king. Hamlet, the prince, ‘caught the conscience’ of the murderer, and was sent off to England. As Hamlet is aware, the companions Claudius assigned him for the journey, Rosencrantz and Guildenstern, bore letters giving the English court a mandate to immediately execute Hamlet. Hamlet, however, intends to turn the tables on them, and have his companions executed on arrival in England. The petard was an early form cannon notorious for exploding, killing its own engineer.
[10] ‘The problem of irony, beyond compare the most profound and most alluring in the world’ Thomas Mann, Bemühungen; quoted in D. J. Enright, The Alluring Problem: An Essay on Irony (Oxford, 1986), vii.
[11] D. C. Muecke, The Compass of Irony (London, 1969), 42.
[12] Ibid .25.

[13] Toia [1982] 1 NZLR 555, 559.
[14] Hill [1996] Crim. LR 419. Cf. Middleton (The Times, 12 April 2000); Ghani (Unreported, CA (Crim Div), 25 October 1999); Harron [1996] Crim. LR 581; Burge & Pegg [1996] 1 Cr App R 163; Higgins & Litchfield (The Times, 11 August 1995); Mazur (Unreported, Vic CCA, 8 June 2000); Bandiera & Licastro (Unreported, Vic CCA, 11 November 1999); Chan (Unreported, Vic CCA, 12 March 1998); Manapouri [1995] NZLR 407 (CA); Samuels [1985] 1 NZLR 350 (CA); Toia [1982] 1 NZLR 555 (CA). It is hard to believe that less than thirty years ago in Chapman [1973] 1 QB 774, 783 the Court of Appeal commented: ‘Curiously enough this point [whether the defendant’s false evidence can constitute corroboration for prosecution evidence] seems never to have arisen for direct decision by the Court of Appeal in this country either in a civil or criminal case.’
[15] Bandiera & Licastro (Unreported, Vic CCA, 11 November 1999); austlii.edu.au. This prosecutorial response is less likely in England, since the trial judge may provide direction on the guilty-lie inference independently of the prosecution case. See below, nn 85-94.
[16] This is a an example of double irony. Compare the opening of Chapter Four of Book Five of Anatole France’s allegorical history, Penguin Island, ‘The Penguins had the finest army in the world. So had the Porpoises’. Muecke suggests that ‘the contradiction and mutual destruction directs us to the ironist’s real meaning.’ Muecke (1969) op. cit. 24. See also, D. C. Muecke, Irony and the Ironic (London, 1970), 34.
[17] Bandiera & Licastro (Unreported, Vic CCA, 11 November 1999); austlii.edu.au.
[18] Ibid.
[19] Ibid.
[20] Middleton (The Times, 12 April 2000); lexis.com; cf. Mazur (Unreported, Vic CCA, 8 June 2000); Chan (Unreported, Vic CCA, 12 March 1998); Perera [1982] VicRp 91; (1982) 6 A Crim R 225, 235 (Murray J). D. Mathias, ‘Lies Directions’ [1995] New Zealand Law Journal 307.
[21] Unreported, Vic CCA, 12 March 1998; austlii.edu.au.
[22] [1997] 2 VR 88.
[23] Compare the earnest tone of G. Lyon and B. Walmsely, ‘Consciousness of Guilt: The use of lies told by the accused’ (Nov 1997) Law Institute Journal 50, 51: [S]uch was the concern at the “developing mystique” surrounding the use of lies in trials, the Court reiterated in clear and unambiguous language a number of the principal considerations on the subject’.
[24] Chan (Unreported, Vic CCA, 12 March 1998); austlii.edu.au; cf. R v Mazur (Unreported, 8 June 2000), (Brooking JA, Chernov JA and Winneke P agreeing).
[25] Many of these were discussed and analysed in Chapman [1973] 1 QB 774 without successful reconciliation.
[26] [1949] AC 253 (PC).
[27] Kerim (1987) 1 QD R 426, 445 (McPherson J ). The catch-22 irony is discussed further in Section 3.4.
[28] As the Court noted in Laz [1998] 1 VR 453, 467 ‘very special problems arise when the same lie and thus the same consciousness of guilt is argued to be applicable to charges arising out of more than one event’. In that case, while the court considered the inference logically open with regard to one of the charges, it would have been so difficult to explain this to the jury that ‘this was not a safe case to let any lie admitted or otherwise go to the jury as amounting to consciousness of guilt and thus an implied admission of one or both of the counts charged.’ (467-8)
[29] Broadhurst [1964] AC 441 (PC), 457 (Lord Devlin): ‘Save in one respect, a case in which an accused gives untruthful evidence is no different from one in which he gives no evidence at all. But if upon the proved facts two inferences may be drawn about the accused’s conduct or state of mind, his untruthfulness is a factor which the jury can properly take into account as strengthening the inference of guilt.’ (Emphasis added.)
[30] Cf. D. Hamer, ‘The Continuing Saga of the Chamberlain direction: Untangling the Cables and Chains of Criminal Proof’ [1997] MonashULawRw 5; (1997) 23 Monash University Law Review 43, 70; D. L. Martin, ‘White and Côté’ : A Case Comment’ (1997) 42 McGill Law Journal 459, 463.
[31] The term ‘chain’ has been used to describe inferences for many years. Early references include Bentham Rationale of Judicial Evidence (1827) 3:3 and Burrill, A treatise on the nature, principles and rules of circumstantial evidence, especially that of the presumptive kind, in criminal cases (1856), 598–600.
[32] Perera [1982] VicRp 91; (1982) 6 A Crim R 225, 238 (Marks J).
[33] Hickey (Unreported, CA (Crim Div), 30 July 1997); lexis.com, quoting from Burge & Pegg [1996] 1 Cr App R 163, 174.
[34] (1987) 1 QD R 426.
[35] Ibid. 449 (Macrossan J).
[36] Ibid. 450 (Macrossan J).
[37] [1993] HCA 63; (1993) 178 CLR 193.
[38] Ibid. 210 (Deane, Dawson and Gaudron JJ). Cf. White and Côté [1998] 2 SCR 72.
[39] [1993] Crim. LR 946; lexis.com.
[40] The classic direction laid down in Lucas (Ruth) [1981] 1 QB 720, 724, distinguishes between the requirement that the lie relate to a material issue, and the requirement that the motivation of the lie be a realisation of guilt. But this appears to be just two ways of stating the same requirement. Materiality seems to be a test for the motivation requirement. If a lie appeared to be motivated by a consciousness of guilt then it could scarcely be described as trivial. If a lie appeared not to be motivated by a consciousness of guilt, then it would not be appropriate to describe it as material.
[41] Ali Ahmed [1993] Crim. LR 946; lexis.com. Cf. Lucas (Ruth) [1981] 1 QB 720; Peacock [1998] Crim. LR 681; Turnbull [1977] 1 QB 224.
[42] Harris (1990) SASR 321, 323 (King CJ); see also Palmer op. cit. 144-5.
[43] Starr v US [1897] USSC 11; 164 US 627 (1896). Discussed in Palmer op. cit. 137.
[44] Palmer op. cit. 137.
[45] [1993] 3 All ER 225.
[46] Ibid. 231.
[47] [1985] 1 NZLR 350.
[48] Ibid. 355.
[49] Ibid.
[50] [1995] Crim. LR 80; lexis.com.
[51] The Times, 12 April 2000; lexis.com.
[52] The Independent, 9 Feb 1987.
[53] Richens [1992] EWCA Crim 3; [1993] 4 All ER 877.
[54] Ibid. 885.Cf. Gordon [1995] Crim. LR 306; Burnham [1995] Crim. LR 491; Tucker [1994] Crim. LR 683.
[55] Palmer op. cit. 110.
[56] (1989) 50 A Crim R 327.
[57] Ibid. 333.
[58] Hickey (Unreported, CA (Crim Div), 30 July 1997); lexis.com (emphasis added); quoting from Burge & Pegg [1996] 1 Cr App R 163, 174; cf. Holman [1995] Crim. LR 80; Ali Ahmed [1993] Crim. LR 946.
[59] Above n 38.
[60] Toia [1982] 1 NZLR 555; see also Mathias op. cit. 311.
[61] Morrison v Jenkins (1949) 60 CLR 626, 644; cf. R. Eggleston Evidence, Proof and Probability (London, 1983), 37–8.

[62] A. Welsh, Strong Representations: Narrative and Circumstantial Evidence in England (Baltimore, 1992), 75.
[63] Zoneff (2000) 172 ALR 1, 19 (Kirby J), quoting from Toia [1982] 1 NZLR 555, 559.
[64] Broadhurst [1964] AC 441, 457 (PC); followed in Dehar [1969] NZLR 763; Samuels [1985] 1 NZLR 350, 355; Bey [1993] 3 All ER 253; [1994] 1 WLR 39; Goodway [1993] 4 All ER 894; Tucker [1994] Crim. LR 683.
[65] Sutton (1986) 5 NSWLR 697, 701; Li Zheng (1995) 83 A Crim R 572, 577; Mercer (1993) 67 A Crim R 91, 96.
[66] Goodway [1993] 4 All ER 894, 901–2; Burge and Pegg [1996] 1 Cr App R 163, 172.
[67] [1994] Crim. LR 683; lexis.com.
[68] The appeal was dismissed by application of the proviso.
[69] Tucker [1994] Crim. LR 683.
[70] Zoneff (2000) 172 ALR 1 (Gleeson CJ, Gaudron, Gummow and Callinan JJ), 7. Kirby J, dissented on this point: ‘The criterion must be the way the jury might use the evidence not the subjective purpose of the prosecutor in eliciting the evidence or relying upon it.’ (at 22)
[71] [1996] 1 Cr App R 163.
[72] Ibid. 172.
[73] Ibid. 173.
[74] Ibid.
[75] Ibid. 174.
[76] Ibid.
[77] [1993] 4 All ER 894.
[78] Burge & Pegg [1996] 1 Cr App R 163, 173.
[79] Goodway [1993] 4 All ER 894, 902.
[80] This would be an irony of event. See below n 94.
[81] Zoneff (2000) 172 ALR 1, 21 (Kirby J).
[82] Indeed, it may be ‘so very relevant that to exclude it would be an affront to common sense.’ Boardman [1975] AC at 456 (Lord Cross); quoted in Pfennig [1995] HCA 7; (1995) 182 CLR 461, 514 (McHugh J).
[83] Boardman [1975] AC 421, 456 (Lord Cross), quoted in Pfennig [1995] HCA 7; (1995) 182 CLR 461, 478 (Mason CJ, Deane and Dawson JJ). See also at 488 (Mason CJ, Deane and Dawson JJ). Cf. Harriman [1989] HCA 50; (1989) 167 CLR 590, 597 (Dawson J), quoted in Pfennig (1995) 182 CLR 46, 501 (Toohey J).
[84] Pfennig [1995] HCA 7; (1995) 182 CLR 461, 512–3 (McHugh J) quoted in Gipp (1988) 194 CLR 107, 134–5 (McHugh and Hayne JJ). McHugh J also notes a number of other policy difficulties with propensity evidence. See also, Gipp (1998) 194 CLR 107, 164–5 (Callinan J) citing BRS [1997] HCA 47; (1997) 191 CLR 275, 305 (McHugh J).
[85] Mercer (1993) 67 A Crim R 91, 96.
[86] Middleton The Times, 12 April 2000.; lexis.com.
[87] The Court cited Smith’s commentary on Smith [1995] Crim. LR 940, and the reference to ‘an inadmissible chan of reasoning’ in Boardman [1975] AC 421, 453 (Lord Hailsham). See also A. Palmer, ‘The Scope of the Similar Fact Rule’ [1994] AdelLawRw 4; (1994) 16 Adelaide Law Review 161, 178-187.
[88] Arcangioli [1994] 1 SCR 129143; quoting from E. W. Cleary et al, eds, McCormick on Evidence 3rd ed (St Paul, 1984), 803; quoted in Martin op. cit., 473.
[89] Chan (Unreported, Vic CCA, 12 March 1998); austlii.edu.au; (Brooking JA) quoting from Coke 3 Inst 232. Coke also suggested such a miscarriage of justice would provide a caveat ‘to judges, that they in case of life judge not too hastily upon bare presumption’.
[90] Muecke (1969) op. cit. 19-20. The difficulty of providing an ultimate definition of irony is acknowledged by many commentators. Eg, ‘Definitions?’, Enright op. cit. 1-7; Muecke (1970) op. cit. 7-8.
[91] In Ali Ahmed [1993] Crim. LR 946; lexis.com it was suggested that proof of the defendant’s lie should come from ‘an admission by the defendant, or by evidence from an independent witness’. Similarly in Liacopoulous (Unreported, CA (Crim Div), 31 August 1994); lexis.com there was reference to the defendant’s lie, proven ‘either because he admits it or because of other extensive evidence’. However, there are other authorities where the falsity of the statement appears from its inherent implausibility. For example, the defendant in Velevski (Unreported, NSW CCA, 10 May 1999); austlii.edu.au claimed to have been asleep in one bedroom while four members of his family were killed in another bedroom. He claimed to have slept for a period of 14 to 15 hours without getting up. The Crown alleged this was a lie ‘as a matter of common sense’, as well as on the basis of expert evidence on sleep patterns. For a discussion of the reasons for rejecting the ‘lower level’ interpretation in verbal ironies, see W. C. Booth’s discussion of ‘Clues to Irony’ in A Rhetoric of Irony (Chicago, 1974), 49.
[92] Other commentators have questioned the need for a victim: Enright op. cit. 152; Booth op. cit. 27-8. In the case of the guilty-lie irony, it may also be questioned whether the defendant is altogether confident that there strategy will work. Nevertheless, if the defendant had foreseen that their ploy would rebound on them, clearly they would not have adopted it.
[93] See above, n 9. For examples of this kind of irony in crime fiction, see D.E. Westlake (ed), Murderous Schemes: An Anthology of Classic Detective Stories (Oxford, 1996) 378-89, 403-54.
[94] See Muecke (1969) op. cit. 32; Muecke (1970) op. cit. 69.
[95] D. N. MacCormick, Legal Reasoning and Legal Theory (Oxford, 1978), 90; cf. B. Jackson, ‘Narrative Models in Legal Proof’ (1988b) I International Journal for the Semiotics of Law 225; Welsh op. cit., 33; R. Allen, ‘The Nature of Juridical Proof’ (1991a) 13 Cardozo Law Review 373; W. Twining, ‘Lawyer’s Stories’, Rethinking Evidence: Exploratory Essays (Oxford, 1990); P. Bergman and A.Moore, ‘Mistrial by likelihood ratio: Bayesian analysis meets the F word’ (1991) 13 Cardozo Law Review 589; N. Pennington and R. Hastie, ‘A Cognitive Theory of Juror Decision Making: The Story Model’ (1991) 13 Cardozo Law Review 519, 520; W. Bennet and M. Feldman, Reconstructing Reality in the Courtroom: Justice and Judgment in American Culture (New Brunswick, 1981); W. A. Wagenaar, P. van Koppen and H. Crombag, Anchored Narratives: The Psychology of Criminal Evidence (Hertfordshire, 1993); M. Redmayne, ‘A Likely Story!’ (1999) 19 Oxford Journal of Legal Studies 659, 660.
[96] Wills on Circumstantial Evidence (6th ed, London, 1912), 48. Cf. Pfennig[1995] HCA 7; , (1995) 182 CLR 461, 536 (McHugh); N. Pennington and R. Hastie ‘Explaining the evidence: Tests of the story model for jury decision-making’ (1992) 62 Journal of Personality and Social Psychology 189, 252; Bennett and Feldman, op. cit. 66–8; Wagenaar et al, op. cit. 141–3, 211, 218; R. Lempert ‘Telling Tales in Court: Trial Procedure and the Story Model’ (1991) 13 Cardozo Law Review 559; R. Hastie (editor), Inside the Juror: The Psychology of Juror Decision Making (Cambridge 1993), 27; Bergmann and Moore, op. cit. 602.
[97] Vendetta (Faber and Faber, London, 1990) 73.
[98] 1980, trans. W. Weaver, Minerva Edition, London, 1992.
[99] Ibid. 470; cf. at 492.
[100] Julian Barnes Flaubert’s Parrot, quoted in Enright, op. cit. 55. Barnes’s narrator wonders whether ‘the wittiest, most resonant irony isn’t just a well-brushed, well-educated coincidence.’ Enright, op. cit. 144, suggests ‘[i]rony may be a last-ditch defence against – what humans cannot endure – meaninglessness, by identifying or postulating a connection or cross-reference and so registering or creating a sort of desperate almost- sense.’
[101] Muecke (1970), op. cit. 47.
[102] Ibid., 69. Cf. Booth, op. cit. 37, 43; Muecke (1969), op. cit. 226; N. Knox ‘On the Classification of Ironies’ Modern Philology (August 1972) 53, 61; ‘Whoever has reached such a standpoint of godlike genius consequently looks down in superior fashion on all other mortals’: Hegel writing of the ironist in Vorlesungen über die Aesthetik; quoted by Enright, op. cit. 8.
[103] Booth, op. cit. 44; Muecke (1969), op. cit. 31.
[104] Van der Meer [1988] HCA 56; (1988) 82 ALR 10, 31.

[105] Joseph Heller, Catch-22 (1961), R. M. Scotto (ed) A Critical Edition (New York, 1973), 45-6. Heller tells the story of Yossarian’s attempt to escape from US military service as a bomber in the Second World War. The central catch-22 was that, if Yossarian was crazy, he was entitled to go home. But if he wanted to go home, then that proved that he wasn’t crazy. Paul Fussell, The Great War and Modern Memory (New York, 1975), 176-7, notes a historical example from the First World War:

One of the highest achievements in saying the thing that is not was the poster announcing the terms of the Military Service Act of 1916: ‘“EVERY UNMARRIED MAN OF MILITARY AGE ... CAN CHOOSE ONE OF TWO COURSES:

(1) He can ENLIST AT ONCE and join the Colors without delay;

(2) He can ATTEST AT ONCE UNDER THE GROUP SYSTEM and be called up in due course with his Group.”

And then the climax, anticipating the world of Heller’s Catch-22:

“If he does neither, a third course awaits him: HE WILL BE DEEMED TO HAVE ENLISTED ...”
[106] ‘This suggestion that somehow he can roll back GST and make it simpler is like saying that he’d like to head in a northerly direction by walking south. It's mutually incompatible, it's inconsistent, it can't be done.’ Peter Costello (Treasurer), commenting on Australian Labor Party policy to partially roll back the Goods and Service Tax, while at the same time making it simpler. (http://www.abc.net.au/7.30/s151283.htm)
[107] Muecke (1969), op. cit. 119.
[108] Ibid. 26-7.
[109] Enright, op. cit. 4.
[110] Booth, op. cit. 178.
[111] [1977] 1 QB 224.
[112] Ibid. 235.
[113] Booth, op. cit. 59 n 14.
[114] (1987) 1 QD R 426. See above n 34.
[115] See (1987) 1 QD R 426, 429, 443, 459.
[116] See Enright, op. cit. 41; Muecke (1970), op. cit. 21.
[117] (1987) 1 QD R 426, 459.
[118] Michael Dibdin hints at this illogicality in Vendetta (op. cit.). His detective, Aurelio Zen, is investigating a murder. The murder weapon appears to be a Remington Shotgun that is missing from the victim, Oscar Burolo’s collection (at 32). Zen favoured the theory that the person behind the murder was Renato Favelloni, and that he had employed a professional killer. The ‘means’ evidence – the use of a gun from the victim’s own collection – seemed superficially inconsistent with this, but Zen, by some circuitous reasoning, managed to make this support his preferred theory. ‘One would of course expect a professional killer to use his own weapon, probably with a silencer. It can be argued that this anomaly merely strengthens the case against Favelloni, indicating that an attempt was made to disguise the fact that the crime was a premeditated conspiracy against the life of Oscar Burolo.’ (at 46)
[119] ‘It is not irony but the desire to understand irony that brings such a chain to a stop.’ Booth, op. cit. 59 n 14.
[120] Edwards [1993] HCA 63; (1993) 178 CLR 193, 211 (Deane, Dawson and Gaudron JJ) quoted and emphasis added in Green [1999] HCA 13; (1999) 161 ALR 648, 651; cf. Edwards [1993] HCA 63; (1993) 178 CLR 193, 198 (Brennan J).
[121] [1981] QB 720.
[122] Ibid. 724.
[123] [1993] HCA 63; (1993) 178 CLR 193.
[124] Ibid. 199 (emphasis added), Brennan J’s proffered solution to this problem is unpersuasive: ‘The relevant inference is not that the accused realised his guilt but that, in making his statement, he was unable to account innocently for the evidence that has been given against him.’
[125] Martin, op. cit. 467. Jackson makes a similar claim when considering ‘whether common sense can ever dictate a direct inference of guilt from ... the equivocal fact of an accused’s failure to testify.’ (J. Jackson, ‘Interpreting the Silence Provisions: The Northern Ireland Cases’ [1995] Crim. LR 587, 600.) He suggests that ‘the court or jury is in a position to draw the ‘proper’ inference from silence only where it knows the reason for silence’, in which case the inference can be no more than ‘an ex posto facto rationalisation of what the trier of fact has already become convinced of’.
[126] Martin, op. cit. 459.
[127] Ibid. [20], quoted by H. Stewart, ‘Towards a Principled Approach to Consciousness of Guilt: A Comment on White and Côté’ (1999), 43 Criminal Law Quarterly 17, 20. See also Zoneff (2000) 172 ALR 1, 7 (Gleeson CJ, Gaudron, Gummow and Callinan JJ), 19 (Kirby J). Cf. Palmer (1998), op. cit. 143: ‘Apart from anything else, it is only after we have decided that the person is guilty that we can confidently say that their behaviour was faked. Until then, the most that can be said is that the behaviour does not ring true.’
[128] Gionfriddo and Faure (1990) 50 A Crim R 327, 339 (Gray J).
[129] Zoneff (2000) 172 ALR 1, 19 (Kirby J), endorsing White & Côté’ (1998) 125 CCC 3d 385, 398.
[130] Edwards [1993] HCA 63; (1993) 178 CLR 193, 199 (Brennan J); 209 (Deane, Dawson and Gaudron JJ); Laz [1998] 1 VR 453, 468-9; Martin, 476; cf. discussion of criminal standard in Part 1.
[131] R. Freidman ‘Assessing evidence’ (1996) 94 Michigan Law Review 1810, 1828. Friedman was not specifically discussing consciousness-of-guilt reasoning, however, he does seem committed to an inherently forward looking propensity view of probability.
[132] Edwards [1993] HCA 63; (1993) 178 CLR 193, 209 (Deane, Dawson and Gaudron JJ), discussing the defendant’s argument.
[133] It is assumed that the probability of a lie given the defendant’s guilt is greater than the probability of a lie given the defendant’s innocence. That is, P(L|G) > P(L|not-G), where ‘L’ stands for the defendant’s lie, and ‘G’ stands for the defendant’s guilt. It follows that the Bayesian likelihood ratio for the defendant’s lie is greater than one. That is, L(L) = P(L|G) / P(L|not-G) > 1. By Bayes’ Theorem, proof of the lie would therefore increase the odds and the probability of the defendant’s guilt. The ‘odds-likelihood’ form of Bayes’ Theorem is O(G|L) = O(G) x L(L).
[134] Edmunds v Edmunds & Ayscough [1934] VicLawRp 23; [1935] VLR 177, 186 (Lowe J); quoted in Edwards [1993] HCA 63; (1993) 178 CLR 193, 201 (Brennan J); Jan Ha Le Trinh (Unreported, Vic CCA, 17 December 1998); austlii.edu.au; austlii.edu.au. There have been cases where courts have entertained the possibility that a defendant’s simple denial could count against them. However, in these cases the inference of guilt was not based on the defendant’s words, but rather ‘the manner and tone of the words used by the accused and the circumstances of their utterance. A man’s looks may belie him. Demeanour and conduct may discount denial and manifest guilt as surely as would a confession made by words.’ Woon [1964] HCA 23; (1964) 109 CLR 529, 541 (Windeyer J), quoted in Palmer (1998), op. cit. 146. See also Christie [1914] UKLawRpAC 20; [1914] AC 545, 554 (Lord Atkinson), 560 (Lord Moulton); R v B (SP) (1994) 90 CCC (3d) 478, 483, discussed in Palmer (1998), op. cit. 147. Palmer questions the stereotypical psychological assumptions underlying the inference. In Christie, Lord Moulton suggested that ‘the evidential value of the behaviour of the accused where he denies the charge is very small either for or against him’ and in most cases should be excluded by the exercise of the trial judge’s discretion. (at 560)
[135] Gionfriddo & Faure (1990) 50 A Crim R 327, 333; Laz [1998] 1 VR 453, 466. Jan Ha Le Trinh (Unreported, Vic CCA, 17 December 1998), [20-1]; Perera [1982] VicRp 91; (1982) 6 A Crim R 225, 235.
[136] The defendant in Meskers (Unreported, NSW CCA, 13 June 1991); austlii.edu.au placed himself in a catch-22. He was appealing against conviction for indecent assault of a girl under 16. He had previously been convicted following a plea of guilty to similar offences against his daughter, a friend of the complainant. As a consequence he was presented with a dilemma. If he tried to keep this history from the jury ‘this would curtail effective cross examination of the victim, and his daughter, each of whom were to be called at the trial, to the effect that they had collaborated, and had a motive to tell lies because of the disputes on foot concerning the marriage and claims to access or custody of the appellant’s children. To introduce, through cross examination of the girls, or otherwise in his own case, any suggestion that the appellant had sexually abused his own daughter, it was argued would produce incurable prejudice; yet, so the submission went, without that emerging, the true background for the complaints and the possibility of a motive in the girls to give false evidence, would remain hidden.’ This led the defendant to seek a stay of prosecution on the ground he could not be given a fair trial. The defendant’s appeal was dismissed. Wood J indicated: ‘The dilemma in which the accused found himself arose because he had earlier been convicted of the sexual assault of his daughter. It would, however, be a very curious [and ironical] result if that fact rendered him immune from prosecution for a separate assault or series of assaults on another victim, or even the same victim. ... The fact that the appellant faced this difficult tactical decision did not, however, involve any fundamental defect in the trial, or involve any inherent unfairness. It arose, as I have observed, as the result of his own earlier misconduct, and not because of anything in the Crown case.’
[137] Muecke (1969), op. cit. 120.
[138] H. J. Stark, ‘Catch-22: The Ultimate Irony’, in J. Nagel-James (ed), Critical Essays on Catch-22 (Encino, 1974), 140.
[139] Muecke (1969), op. cit. 26.
[140] R. Brustein, ‘The Logic of Survival in a Lunatic World’ in Nagel-James (ed) op. cit., 17.
[141] The tone, however, was caught by the trial judge in Whitby [1977] 1 QB 224. The defendant provided an alibi from his wife. The trial judge virtually said, ‘well he would, wouldn’t he?’ (See below, n 150). ‘[W]hat can the poor wife do other than back her husband up as much as she possibly can.’ The Court of Appeal considered this kind of reasoning unfair and absurd, pointing to the dilemma that it poses for the defendant. ‘[A]s the appellant’s alibi was that he was at home with his wife, who else could he have called to support it?’
[142] Eco, op. cit. 372. Remigio did confess. As it happens, he was ‘guilty’ of being a Pseudo Apostle. However, he was also induced to confess to the murders that were occurring at the Abbey, of which he was innocent.
[143] Ibid. 381.
[144] D. Parker ‘Answering Idle Questions: Open and Closed Readers’ in M Thomas Inge (ed) Naming the Rose (Jackson 1988), 152.
[145] T. Colletti, Naming the Rose: Eco, Medieval Signs and Modern Theory (Ithaca, 1988). This is probably a coincidence rather than an irony. Cf. Enright, op. cit. 55. See above, n 100.
[146] Colletti, op. cit. 99-100
[147] Ibid. 101.
[148] Dibdin, op. cit. 26.
[149] Ibid. 112.
[150] Ibid. 222. Dibdin may have borrowed from Mandy Rice-Davies. On 29 June 1963, while in the witness box in the trial of Stephen Ward, she was told that Lord Astor denied her allegations regarding him and his house parties at Cliveden. She responded, ‘He would, wouldn’t he?’ Oxford Dictionary of Quotations (5th ed, Oxford, 1999) 626:9; The Guardian (Manchester, 1 July 1963).
[151] Third Floor Productions. For further discussions of the case and the film see: R. K. Sherwin, ‘Law Frames: Historical Truth and Narrative Necessity in a Criminal Case’ (1994) 47 Stanford Law Review 39; M. Brower, ‘Crossing a line that is not thin at all, Randall Dale Adams wins release from a Texas Prison’ People Weekly (10 April 1989), 155; F. Bruning, ‘Why did Randall Adams almost die?’ Maclean’s (27 March 1989), 9.
[152] The case of Stephen Downing presents a recent example of the guilty-lie irony at sentencing. It was not enough that Downing was wrongly convicted. ‘He was forced to spend 10 years more than the 17-year sentence handed down by the courts in 1973 because his refusal to confess made him ineligible for parole.’ The Observer (London, 18 February 2001), 1.
[153] Kennion (Unreported, CA (Crim Div), 16 Jan 1997); lexis.com Liacopoulos (Unreported, CA (Crim Div), 31 August 1994); lexis.com; Burge & Pegg [1996] 1 Cr App R 163, 173.
[154] Robinson [1996] Crim. LR 417; lexis.com; Liacopoulous (Unreported, CA (Crim Div), 31 August 1994); lexis.com; Burge & Pegg [1996] 1 Cr App R 163, 173; Hill [1996] Crim. LR 419; lexis.com; Topham (Unreported, CA (Crim Div) 19 March 1995); lexis.com.
[155] Perera [1982] VicRp 91; (1982) 6 A Crim R 225, 236 (Murray J).
[156] Landon [1995] Crim. LR 338; lexis.com.
[157] Peacock [1998] Crim. LR 681; lexis.com.
[158] Higgins & Litchfield (Unreported, CA (Crim Div) 16 June 1995); lexis.com.
[159] Peacock [1998] Crim. LR 681; lexis.com; Harron [1996] Crim. LR 581; lexis.com; Landon [1995] Crim. LR 338; lexis.com.
[160] Robinson [1996] Crim. LR 417; lexis.com; Topham (Unreported, CA (Crim Div) 19 March 1995); lexis.com; Burge & Pegg [1996] 1 Cr App R 163, 173.
[161] Topham (Unreported, CA (Crim Div) 19 March 1995); lexis.com; Kennion (Unreported, CA (Crim Div), 16 Jan 1997); lexis.com.
[162] Perera [1982] VicRp 91; (1982) 6 A Crim R 225, 236 (Murray J).
[163] Robinson [1996] Crim. LR 417; lexis.com; Liacopoulos (Unreported, CA (Crim Div), 31 August 1994); lexis.com; Hill [1996] Crim. LR 419; lexis.com.
[164] Topham (Unreported, CA (Crim Div) 19 March 1995); lexis.com; Burge & Pegg [1996] 1 Cr App R 163, 173.
[165] Higgins & Litchfield (Unreported, CA (Crim Div) 16 June 1995); lexis.com. The Court in this conspiracy and attempted murder case extended the concept of ‘central issue’ to its limits. A relatively clear application concerned the question of the truthfulness of the defendants’ claims that the victim was uninjured when they left him on the night of the attack. However, the principle was also applied to ‘six ancillary matters’, including statements relating to business between the victim and the defendants months before the attack. The Court’s reasoning was that these ‘were not individually or collectively separate matters from which the jury could infer guilt. They were part of the whole prosecution case on the charge of conspiracy, part of the “scene-setting” ... The jury’s view of them was likely to stand or fall with their decision on that main issue.’
[166] Unreported, CA (Crim Div), 31 August 1994; lexis.com.
[167] [1994] Crim. LR 682; lexis.com.
[168] [1995] Crim. LR 80; lexis.com.
[169] [1993] 4 All ER 894.
[170] [1969] NZLR 763.
[171]Goodway [1993] 4 All ER 894, 901; Dehar [1969] NZLR 763, 765.
[172] [1993] 4 All ER 894, 902.
[173] See above nn 153–65.
[174] Dehar [1969] NZLR 763 765; affirmed in Goodway [1993] 4 All ER 894, 901; House & Meadows [1994] Crim. LR 682; Holman [1995] Crim. LR 80; Burge & Pegg [1996] 1 Cr App R 163, 172.
[175] Peacock [1998] Crim. LR 681; lexis.com (emphasis added).
[176] Barsoum [1994] Crim. LR 194; lexis.com (emphasis added).
[177] Bartlett (Unreported, CA (Crim Div), 23 October 2000); lexis.com; cf. Leeson [2000] Crim. LR 195. It seems that the application of the central-issue principle in Bartlett is questionable. The complainant was in a dispute with the defendant and her husband, a co-defendant, as to who was entitled to possession of a flat. The complainant was in current possession, although the defendant’s husband also claimed to be spending time there. The defendant and co-defendant broke in to the flat and took certain items of the complainant including a camera, and were convicted of theft. On being found in possession of the camera, the defendant initially said to police ‘I’ve got a receipt for that’. However, she was unable to produce the receipt, and later changed her story, claiming that the camera had been taken by mistake and that she intended to return it. The prosecution were, therefore, alleging that the defendant had told two lies with respect to the camera, first, that it was hers, and secondly, that she had taken it by error and planned to return it. The first, having been abandoned by the defendant, was not at the heart of her defence at trial, and may have supported the guilty-lie inference.
[178] Bartlett (Unreported, CA (Crim Div), 23 October 2000); cf. Leeson [2000] Crim. LR 195. It seems that the application of the central-issue principle in Bartlett is questionable. The complainant was in a dispute with the defendant and her husband, a co-defendant, as to who was entitled to possession of a flat. The complainant was in current possession, although the defendant’s husband also claimed to be spending time there. The defendant and co-defendant broke in to the flat and took certain items of the complainant including a camera, and were convicted of theft. On being found in possession of the camera, the defendant initially said to police ‘I’ve got a receipt for that’. However, she was unable to produce the receipt, and later changed her story, claiming that the camera had been taken by mistake and that she intended to return it. The prosecution were, therefore, alleging that the defendant had told two lies with respect to the camera, first, that it was hers, and secondly, that she had taken it by error and planned to return it. The first, having been abandoned by the defendant, was not at the heart of her defence at trial, and may have supported the guilty-lie inference.
[179] Fitzgibbon, The Independent (3 September 1993); lexis.com.
[180] Middleton The Times, 12 April 2000; lexis.com.
[181] The Times, 12 April 2000; lexis.com.
[182] Ibid. (emphasis added).
[183] [1969] NZLR 763.
[184] [1993] 4 All ER 894; above nn 171–2.
[185] Middleton The Times, 12 April 2000; lexis.com.
[186] Ibid.
[187] Ibid.
[188] Ibid.
[189] (1995) 83 A Crim R 572.
[190] Ibid. 576-7.
[191] Ibid. Cf. Laz [1998] 1 VR 453, 466.
[192] Unreported, Vic CCA, 17 December 1998; austlii.edu.au.
[193] Above n 20.
[194] Enright, op. cit. 8-9, quoting from Kierkegaard’s The Concept of Irony: With Constant Reference to Socrates.