skepticlawyer

It is not a private citizen’s duty to prove his innocence, it is the government’s to prove his guilt.

- Ian Welsh

I’ve selected the above quotation because it illustrates in stark form the power differential between the typical accused and the typical prosecuting authority. The alleged criminal has arrayed against him the full coercive powers of the state. For all the common gripes about an under-funded DPP/CPS or overstretched police force, taken together – along with the institutional state apparat of the courts – the power inequality is almost Kafkaesque.

While not all people are alert to the notion that a jury never finds anyone ‘innocent’, most people – even non-lawyers – can recite the tried and true common law principle that the accused is presumed innocent. While media bodies periodically need to be reminded that incautious reporting of allegations (mainly by incautiously forgetting the word ‘alleged’ or by ‘helpfully’ photographing crime scenes) can have deeply prejudicial effects on the accused’s prospects at trial, a lawyer or law student is often needed to remind ordinary folk of what the presumption means at law.

In its simplest form, the ‘presumption of innocence’ means that the Crown, State, Prosecution or what-have-you bears the persuasive burden of proof to a high standard – beyond reasonable doubt. Judges are enjoined against explaining what this means to juries during summing-up, and attempts to do so have more often than not resulted in complex and costly appeals. The current Queensland Supreme Court Benchbook advises its justices to instruct juries only that the standard is high, but is to be interpreted using the words’ ordinary and natural meaning.

In practice, the ‘presumption of innocence’ is a series of interlocked sub-rules, and in some ways it’s better to consider it a surrogate for the allocation of the persuasive burden of proof. The right to silence – both pre- and during trial, is included in this ‘box of rules’. As I discuss later, recent changes mean considerably more ‘burdens’ for the accused under British legislation.

What many people don’t realise is that exceptions to this presumptive rule have been enshrined at common law from the get-go. Even Woolmington, the case that gave English law one of its most memorable expressions – ‘throughout the web of the English criminal law one golden thread is always to be seen, that it is the duty of the prosecution to prove the prisoner’s guilt’ – provided for exceptions based on insanity and statute. The latter have proliferated in recent times, and are central to my discussion here.

The presumption of innocence, then, puts the Crown to the proof, and does so to a high standard. It does this in large part because the state is strong, while the individual is weak. In common parlance, it’s meant to give every mug an even break. It’s not always popular – there are times when I suspect the only people who believe in it are lawyers – but the state’s task is difficult for a reason. The consequences for individuals traduced by the criminal law are very serious indeed.

These days, the presumption of innocence is increasingly abrogated, both through statutes creating strict liability regimes – particularly for summary offences – and the imposition of persuasive burdens on the accused at trial. This latter is often coupled with a watering down of the right to silence, also by statutory intervention. The law is unsettled, and it’s possible to argue that recent rescue attempts under the European Convention on Human Rights and the 1998 Human Rights Act (UK) have created more confusion than clarity.

Australian and British law have also bifurcated dramatically on this point, while constitutional protections enshrined in US law – memorably assayed by the line beloved of Congressmen in a tight corner (‘I’ll take the Fifth on that’) – mean there is little real harmony across related jurisdictions.

‘Reverse onus’ provisions, in one sense, are true to label. They place burdens on the accused, often with the purpose in mind of lightening the state’s load. They’re meant to make the Crown’s job that much easier. Interestingly, at common law, they tend to operate at extremes of the criminal scale – even descending into tort. In Canada, for example, a driver who hits a pedestrian must prove that he wasn’t driving negligently. Section 101 of England’s Magistrate’s Courts Act provides that a defendant relying on a whole gamut of common law defences must prove his defence. This is essentially the same as strict liability – which tends to attach itself to traffic matters, for example – offences of which type, according to Andrew Ashworth, now form the bulk of the criminal law.

At the other extreme is s 11(2) of the Terrorism Act 2000 (UK). This requires – in order to enliven the defence that one joined a proscribed organization before its proscription, and was inactive thereafter – persuasive proof. I’ll return to the more serious offences later, and treat them in light of ss 34 and 35 of the Criminal Justice and Public Order Act (UK), which involves other derogations separate from the presumption of innocence.

Of course, the defendant’s task is different – he must make out his defence under reverse onus provisions on the balance of probabilities, the civil standard. Nonetheless, we’re still dealing with a persuasive burden, not a mere evidential one. The distinction is important, and was first identified by Thayer, the great 19th century evidence scholar. Thayer noted that M’Naughten, the case that gave the common law its insanity rules (including the allocation of a persuasive burden to the accused) clearly contemplated a persuasive burden, although it was not labelled as such.

A persuasive burden requires proof to the requisite standard. In a civil case, this is a balancing of possibilities that considers whether one possibility is more likely than not. Under the Woolmington rule (by way of contrast), the accused had only to adduce evidence sufficient to raise an issue on the facts (say, self-defence or provocation) and so rebut the Crown case (which of course had to be made out beyond reasonable doubt). As part of the process, the Crown must negative the accused’s evidence. Any statutory exceptions, of course, had to abrogate the presumption plainly, as – to cite a recent example – in R v Director of Serious Fraud Office, ex parte Smith. Insanity’s modern cousin is diminished responsibility, a ‘temporary disease of the mind’ that ‘substantially impairs’ responsibility. It applies only to murder, and like insanity, places a persuasive burden on the accused.

What, then does the rule in Woolmington permit? It seems to allow explicit statutory derogation from the presumption, but what about other tinkering at the margin?

The House of Lords and Court of Appeal attempted to clear up the conundrum in two cases, R v Edwards and R v Hunt. Both concerned charges that fell well down the totem pole of criminal seriousness: Edwards was accused of selling alcohol without a licence, while Hunt was alleged to be in possession of a prescription drug. Both cases involved legislation that asked the court to consider whether there is any such animal as an implied statutory exception to the rule in Woolmington.

Part of the problem in Edwards – and the significance of this will become clear later – was that the accused elected not to give evidence. The Crown had considerable evidence of liquor sales, but called none on the crucial issue of whether he actually possessed a licence. Ostensibly, the only person able to give evidence on the licence point was Edwards himself, and he did not. That said, the Crown had access to the register of licences and did not call evidence to prove absence of a licence. Edwards made a no-case submission.

Consistent with the nature of the statute, Lawton LJ noted that exceptions like this are ‘limited to offences arising under enactments which prohibit the doing of an act save in specified circumstances or by persons of specified classes or with specified qualifications or with the licence or permission of specified authorities’. It was thus the court’s role to construe the statute, which allows consideration of its purposive intent. Hunt followed Edwards closely, in part – according to Patrick Healy – to avoid dealing with the fall-out attached to constraining parliament’s ability to add freely to the growing mass of regulatory offences.

It’s also worth pointing out that Edwards could opt for a summary trial or trial on indictment – and, understandably, he plumbed for trial on indictment. In economic terms, the latter is a ‘high risk, high payoff’ exercise. By putting the Crown to the proof before a jury, there’s a real risk that conviction will lead to a harsher sentence than simply pleading to a summary charge before a judge alone. At least, however, there is a real choice: trial on indictment in circumstances like that in Edwards could allow an accused to walk free. The real danger in Lawton LJ’s reasoning is that, if the courts construe evidentiary rules for trial on indictment as the legislation demands for summary matters, the accused’s ability to choose is rendered nugatory.

Despite significant scholarly criticism, at least Edwards and Hunt allowed English courts to do what they do best: construe statutes and divine parliament’s intent. At this point, two developments added considerable complexity to the mix.

Sections 34-37 of the Criminal Justice and Public Order Act (UK) allow – under various circumstances – adverse inferences to be drawn from the accused’s silence, both during police questioning (s34) and at trial (s35). In these days of rapid justice and supervised case lists, trials before juries are the exception, not the rule, and the facts in issue are likely to be serious. The legislation isn’t explicitly intended to reverse the onus, unlike the licencing provisions considered in Edwards. What it does do, however, is vest very considerable discretion in the judge to provide directions while summing-up that the accused may only avoid by going into the witness box.

To an Australian, sections 34-37 amount to a ‘statutory Weissensteiner’. In response to the peculiar facts of that case, the High Court held that since the accused was the only person in a position to know what happened to the missing couple, the jury could draw a negative inference from his failure to testify. Since Weissensteiner, the court has retreated from such an explicit derogation from the right to silence. In the 2002 case of R v Dyers, for example, Dyers was accused of child sexual assault. He maintained an alibi, but never called his alibi witness. The trial judge told the jury that, whilst they could not speculate about what that witness might have said, they could assume that if she could have helped Dyers’ case, he would have called her. The High Court held that this direction to the jury interfered with Dyers’ right to silence.

Dyers comes at the end of a series of similar Australian decisions, all of which said that the right to silence is unconditional, and the way to protect it is by directions to a jury. Weissensteiner has never been overruled, however, just distinguished on its facts. The right to silence is an important part of the presumption of innocence – meant to prevent the state from using its coercive power to force the accused into the box – and thereby incriminate him. That said, there are times when the accused does have peculiar – even unique – knowledge, and failure to testify could be made the subject of legitimate comment, at least by counsel. Section 20 of the Commonwealth Evidence Act 1995 goes so far as to permit judicial comment in these circumstances, but does not permit directions to the jury. The fine distinction between ‘comment’ and ‘direction’ has already led to one tremendously complex High Court case – 2001’s Azzopardi v R, which turned on appropriate phraseology.

Where this is complicated for the UK is in provisions like s 11(1) of the Terrorism Act. Arguably, it’s only a matter of time before the UK’s statutory Weissensteiner and the reverse onus provision under the section (or others like it) meet in the middle and completely abrogate the presumption of innocence. Attorney-General’s Reference (No 4 of 2002) foreshadows what it may look like, at least in part.

However, the passage of the 1998 Human Rights Act may constitute a prophylactic against this scenario, in that Article 6(2) of the European Convention on Human Rights protects the presumption, while s 3 of the Act allows any legislative abrogation to be ‘read down’ by the judge in the instant case. This places a tremendous burden on the individual judge, and makes it very difficult to articulate a coherent set of guiding principles for Weissensteiner-like situations. It does seem that the typical judicial response is to tolerate legislative derogations from the presumption in the case of summary and regulatory offences, but to read it down when the allegations are more serious. A consequence of the Criminal Justice and Public Order provisions has been a proliferation of lengthy summings-up, even in relatively simple cases.

However, there is no doubt that Parliament explicitly intended to reverse the onus in the Terrorism Act, allocating a persuasive burden to the accused. The Court vitiated this clear intent on the part of the legislature in a species of judicial legislation unusual in a jurisdiction where parliament is so thoroughly supreme. In a sense, European Union instruments are now being used as a de-facto written constitution, allowing a power of review to the judiciary more familiar to North Americans and Australians, with our written constitutions.

The law is at best unsettled, and the judicial response to the latest outbreak of what Andrew Ashworth calls ‘popular punitivism’ seems to be to use the EU instruments in order to preserve common law presumptions. Those instruments’ origin in the civilian system tends to generate more heat than light, however, and irritates the politicians who write the statutes. The latter may, however, be a good thing.