The Reformulated Contextual Truth Defence: More Radical That First Appears

The Defence of Justification

At common law, defendants have a defence of justification (truth) if they prove, on the balance of probabilities, that the defamatory imputations complained of by the plaintiff and found to be carried by a publication are substantially true. ¹ There is also a statutory defence of justification in s 25 of the Defamation Act 2005. ² Section 25 does not exclude the operation of the common law form of the defence, ³ but for all practical purposes, they overlap.

The rationale underlying defences of justification is that liability in defamation law should only attach where a deserved reputation has been damaged, on the theory that where the truth is told, reputations either cannot be damaged, or are only brought down to their proper level. ⁴

Subject to the matters addressed in the next two sections of this article, if a plaintiff complains that a publication carries more than one imputation, the defendant must prove each of those imputations to be substantially true in order to make out make good a complete defence. ⁵ Where a defendant proves the truth of some, but not all, defamatory imputations complained of by the plaintiff (i.e. partially justifies the defamatory imputations), the plaintiff is entitled to damages in respect of the imputations the defendant fails to prove to be true, unless the defendant can make out some other defence in respect of those imputations. Partial justification by a defendant can operate to mitigate damages, in some cases entirely ⁶ or ‘almost to vanishing point’. ⁷

These principles have been modified in two circumstances: one at common law, the other by the defence of contextual truth in s 26 of the Defamation Act 2005.

Common Law: Common Sting and Nuance Imputations

At common law, defendants may, in limited circumstances, be permitted to rely on and justify, by way of complete defence, an imputation carried by a publication which differs from, but has a common sting with, an imputation complained of by the plaintiff.

In England and Wales, the latitude afforded to defendants is governed by what has come to be known as the Polly Peck principle. ⁸ The classic statement of the Polly Peck principle is as follows:

In cases where the plaintiff selects words from a publication, pleads that in their natural and ordinary meaning the words are defamatory of him, and pleads the meanings which he asserts they bear by way of false innuendo, the defendant is entitled to look at the whole publication in order to aver that in their context the words bear a meaning different from that alleged by the plaintiff. The defendant is entitled to plead that in that meaning the words are true and to give particulars of the facts and matters upon which he relies in support of his plea, as he is required to do by R.S.C., Ord. 82. It is fortuitous that some or all of those facts and matters are culled from parts of the publication of which the plaintiff has not chosen to complain.

Khashoggi v IPC Magazines Ltd (‘Khashoggi’) ¹⁰ affords an example of the operation of the Polly Peck principle. The plaintiff complained that a magazine article had falsely accused her of committing adultery with the president of a foreign nation who was a friend of her husband. The defendants were permitted to plead in answer that the real sting of the article, taken as a whole, was that the plaintiff was promiscuous generally and that that sting was true.

The Polly Peck principle was assumed to be good law in Australia until the turn of the century. ¹¹ In Chakravarti v Advertiser Newspapers Ltd (‘Chakravarti’), however, Brennan CJ and McHugh J said, in obiter, that there was a ‘fundamental defect in the reasoning’ ¹² underlying the principle, namely, that it is ¹³ :

[8] … contrary to the basic rules of common law pleadings and in many contexts will raise issues which can only embarrass the fair trial of the action. Leaving aside technical pleas such as pleas in abatement, defences are either by way of denial or confession and avoidance. A defence which alleges a meaning different from that of the plaintiff is in the old pleading terminology an argumentative plea of not guilty. Under the principles of pleading at common law, it could tender no issue and would be struck out as embarrassing. Under the modern system, articulating an alternative meaning could conceivably make explicit the ground for denying a pleaded imputation. But it would be only in such a case that a defendant's plea of a new defamatory meaning might be supportable as a plea which prevents the plaintiff being taken by surprise. A plea of justification, fair comment or qualified privilege in respect of an imputation not pleaded by the plaintiff does not plead a good defence. It is immaterial that the defendant can justify or otherwise defend the meaning which it attributes to the publication. In our view, the Polly Peck defence or practice contravenes the fundamental principles of common law pleadings. In general it raises a false issue which can only embarrass the fair trial of the actions.…

That view was not endorsed by the remaining members of the court. ¹⁴

In the twenty or so years since Chakravarti, Australian first instance and intermediate appellate courts, drawing upon the obiter view expressed by Brennan CJ and McHugh J, have come to different conclusions about the extent to which defendants in defamation actions may plead and seek to justify imputations in terms that differ from those pleaded by the plaintiff.

In New South Wales, having regard to that state’s ‘rules with respect to pleadings’, defendants are not permitted to plead a defence of justification at common law or under s 25 of the Defamation Act 2005 (NSW) other than to an imputation pleaded by the plaintiff. ¹⁵

In other Australian jurisdictions, as a result of the line of authorities commencing with the decision of the Victorian Court of Appeal in David Syme & Co Ltd v Hore-Lacy (‘Hore-Lacy’), ¹⁶ defendants may plead and seek to justify an imputation that is not substantially different from, and no more serious than, an imputation complained of by the plaintiff. ¹⁷ The rationale underlying the Hore-Lacy line of authorities is that a plaintiff is entitled to a verdict on any pleaded imputation which is not the subject of a successful defence, or any meaning the tribunal of fact would give the publication where that meaning ‘was only a nuance or variant, not substantially different or more serious from that proposed by the plaintiff’. ¹⁸ By parity of reasoning, the Hore-Lacy line of authorities permits defendants to plead and seek to justify any such nuance or variant. ¹⁹

Obviously enough, the Hore-Lacy line of authorities is inconsistent with the Polly Peck principle as it operates in England and Wales. Khashoggi, for example, would be decided differently. Imputations are substantially different from one another if ‘the defendant would have been entitled to plead a different issue, adduce different evidence or conduct the case on a different basis … or possibly whether the justification would be substantially different … ’. ²⁰ On that test, an imputation of general promiscuity is clearly not a nuance or variant of a specific imputation of adultery; the evidence proving the truth of each imputation is necessarily different. An imputation of general promiscuity may also be more serious than a specific imputation of adultery.

In other respects, the scope of the latitude afforded to defendants by the Hore-Lacy line of authorities is far from clear. The prevailing, but not unanimous, view in the cases, for example, is that where a plaintiff complains that a publication imputes that they are guilty of some criminal or other misconduct (a guilt imputation), the defendant may not plead and seek to justify an imputation to the effect that there are reasonable grounds to believe that the plaintiff is so guilty (a reasonable grounds imputation). The rationale for that view is that a reasonable grounds imputation, while less serious than a guilt imputation, is substantially different from it. ²¹

I have long posited, ²² with the support of some distinguished academic colleagues, ²³ but without gaining any traction in the courts, that contrary to the post-Chakravarti jurisprudence, the Polly Peck principle is sound and ought to be reinstated in Australia, because the focus on imputations complained of by the plaintiff that is mandated by that jurisprudence has distorted the law of defamation in three principal ways.

First, because most imputations are false innuendos, ²⁴ experienced pleaders can influence the outcome of defamation actions by the tactical choices they make in their formulation. ²⁵ Well-advised plaintiffs exploit the Hore-Lacy line of authorities by drafting false innuendo imputations with an eye to cutting off avenues of defence that would otherwise be available to defendants, and that would still be available to defendants in England and Wales.

One example concerns cases like Khashoggi, where a publication simultaneously carries a false imputation of specific misconduct and a true imputation of general misconduct with a common sting. Where the plaintiff pleads only a specific imputation, it is not open to the defendant to plead and seek to prove a defence of justification at common law or under s 25 of the Defamation Act 2005 in reliance on a more general imputation with a common sting. In such a case, the defendant would not have any defence of justification at common law or under s 25 in respect of the specific imputation and would be prevented from adducing evidence of the substantial truth of the publication as a whole at trial; and the plaintiff would succeed at trial and have damages assessed without regard to any such evidence.

In Chakravarti, Brennan CJ and McHugh J suggested, by reference to a case with quite different facts, ²⁶ that such an outcome would be sound, because,

it would be outrageous if the defendant could obtain a finding that the article was true in substance and in fact when it plainly was not. ²⁷

That statement was no doubt correct in respect of the example given by their Honours. But, as O’Connor LJ said in Polly Peck, there are questions of fact and degree in every case, and courts always retain the ability to ensure that defendants do not act oppressively. ²⁸ On the facts in Khashoggi, if it could be proved that the plaintiff was generally promiscuous, then it would surely have been a matter for the tribunal of fact to assess whether the article as a whole was true in substance and in fact. It is difficult to sustain any blanket proposition to the effect that there would be something ‘outrageous’ about a finding to that effect. Rather, it might be said that, in such a case, it would be outrageous if the plaintiff could obtain a defamation verdict and have damages assessed without regard to the substantial truth of the publication as a whole.

Another example concerns cases where a publication is capable of imputing both guilt and reasonable grounds imputations.

In such cases, defendants can, but are not obliged to, plead guilt and reasonable grounds imputations as alternatives, to hedge against the risk that the tribunal of fact might find guilt, but not reasonable grounds, imputations to be conveyed, or vice versa. ²⁹ If they do so, a defendant who is unable to prove the truth of a guilt imputation may seek to prove that the reasonable grounds imputation is substantially true. If the tribunal of fact finds that the publication carries the reasonable grounds imputation, and that the defendant has proved the substantial truth of that imputation, then the defendant will have a complete defence. If the tribunal of fact finds that the publication carries a guilt imputation, evidence adduced by the defendant in support of a defence of justification directed to a reasonable grounds imputation will not afford a defence but will be able to be relied on by the defendant in mitigation of damages.

If, however, a plaintiff pleads only a guilt imputation, on the prevailing view in the Australian authorities, the defendant may not plead and seek to justify a reasonable grounds imputation with a common sting. In such a case, the defendant will not be permitted to adduce evidence that there were reasonable grounds to believe that the plaintiff was guilty of criminal or other misconduct. If the plaintiff succeeds, damages will be assessed as if no such grounds existed.

The minority view in the authorities — which I favour — is that this approach distorts defamation law, by handing unfettered control of the imputations to be determined at trial to the plaintiff; and depriving the defendant of the ability to adduce evidence at trial that could establish the substantial truth of the imputations that may in fact be found to be carried by the impugned publication and, if the defence fails, be taken into account in mitigation of damage. ³⁰

Second, by s 8 of the Defamation Act 2005, which reflects the common law position, the cause of action for defamation is the publication of defamatory matter about a plaintiff, not — as was the position in NSW under s 9(2) of the Defamation Act 1974 (NSW) ³¹ — the publication of an imputation.

There is thus a tension between, on the one hand, the focus on the imputations pleaded by the plaintiff in the post-Chakravarti authorities (and, for that matter, the terms of s 25 itself) and, on the other, the nature of the cause of action for defamation mandated by s 8 and the common law.

The focus of justification defences in Australian defamation law has become whether the defendant can prove the substantial truth of the imputations pleaded by the plaintiff, rather than whether they can establish the substantial truth of the imputations found by the tribunal of fact to be carried by the defamatory matter. ³² This leads to a preternatural obsession in Australian defamation law with the imputations drafted by the plaintiff’s advisers. It distracts attention from what, in my view, a defamation action should be about: the balance between the plaintiff’s right to reputation and the defendant’s right to freedom of expression, by reference to what was in fact conveyed by the published words.

Third, by focussing in a strict manner upon the common law rules of pleading, the post-Chakravarti authorities have adopted an old-fashioned view of the identification of issues in adversarial litigation that is inconsistent with modern practice ³³ and the nature of a cause of action that turns upon the meaning of words which are inherently susceptible to different meanings.

The authorities mean that different outcomes can result in defamation actions, on the same facts, dependent wholly upon pleading decisions made by the plaintiff and outside the control or influence of the defendant. The imbalance inherent in the post-Chakravarti authorities generally favours plaintiffs, but not universally: a poor pleading decision by a plaintiff can lead to the loss of a case that would have been won had a different imputation, with the same common sting, been pleaded. ³⁴ I would argue that it ought not be countenanced, in modern litigation, that different outcomes can occur on the same facts based upon a matter entirely within the control of only one party. ³⁵

The potential injustice occasioned by the post-Chakravarti authorities is overcome by the operation of the Polly Peck principle, which acknowledges the reality that publications will very often be capable simultaneously of conveying different imputations and nuances of imputations with a common sting; reduces the risk of damages being assessed in blinkers; and mitigates the often decried criticism that the ‘over-nice attention to the pleading of imputations’ has led to Australian defamation law becoming ‘extremely convoluted and unacceptably confusing’. ³⁶

There is no evidence of which I am aware that the Polly Peck principle has led to ‘outrageous’ outcomes of the kind to which Brennan CJ and McHugh J averted in Chakravarti: indeed, in the one example they gave, the defendant failed. ³⁷ As O’Connor J said in Polly Peck, such outcomes can in any event be avoided by the application of ordinary case management principles directed at ensuring that parties do not act oppressively. ³⁸

Contextual Truth

Before the commencement of the uniform defamation legislation in 2006, statutory defences of contextual truth operated under the defamation laws of New South Wales and Tasmania, ³⁹ but not the other States and Territories. Section 16 of the Defamation Act 1974 (NSW) was in the following terms:

1. Where an imputation complained of is made by the publication of any report, article, letter, note, picture, oral utterance or other thing and another imputation is made by the same publication, the latter imputation is, for the purposes of this section, contextual to the imputation complained of.

The operation of the defence is best illustrated by simple example. Suppose that a publication carries three imputations: that the plaintiff is a murderer; the plaintiff is a thief and the plaintiff is a jay-walker. All three imputations are defamatory, because they impute criminal misconduct by the plaintiff, but obviously, the first two imputations are more serious than the third. Suppose that the defendant could establish the substantial truth of the first two imputations, but not the third. In such a case, a defence of contextual truth would likely succeed, because the reputation of a person who has been correctly branded a murderer and a thief is unlikely to be further injured by being incorrectly branded a jay-walker. ⁴¹

A defence of contextual truth was prescribed by s 26 of the Defamation Act 2005 (NSW). As enacted, it was in the following terms:

It is a defence to the publication of defamatory matter if the defendant proves that—

In Kermode v Fairfax Media Publications Pty Ltd, ⁴³ Simpson J identified a serious drafting deficiency in s 26 that undermined its efficacy. She held that the words ‘in addition to’ in s 26(a) prevented a defendant from relying, as a contextual imputation, on an imputation that had been pleaded by the plaintiff; that is, in the odd jargon of this branch of the law, from ‘pleading back’ such an imputation in support of a defence of contextual truth. Simpson J observed that no corresponding words had appeared in s 16 of the Defamation Act 1974 (NSW).

Simpson J’s view was that the capacity to ‘plead back’ was integral to the objective and operation of a contextual truth defence, because it enabled the ‘ultimate question’ to be put to the jury:

… of all those imputations conveyed, and defamatory, and having regard to those proven to be true: was the plaintiff’s reputation further injured by those imputations of which the defendant had not proven truth? ⁴⁴

That objective, in Simpson J’s view, was fatally undermined by the inclusion of the words ‘in addition to’ in s 26(a) of the Defamation Act 2005, because any imputation that had been pleaded by the plaintiff could not be a contextual imputation for the purposes of the defence, even where the defendant could prove or had proven its truth.

The implications are significant. Returning to the example I gave earlier, suppose a publication imputed, correctly, that the plaintiff was a murderer and a thief, but incorrectly that they were a jay-walker. If the plaintiff sued only on an imputation of jay-walking, imputations of murder and theft would be ‘in addition to’ the imputation of which the plaintiff complains, and thus capable of being relied on as contextual imputations for the purposes of a s 26 defence. The defendant would likely have a good defence in respect of the whole of the publication, with the result that the plaintiff’s action would fail in its entirety, ordinarily with costs.

Suppose, however, that the plaintiff instead sued on all three imputations. In those circumstances, the defendant could not ‘plead back’ any of them, and so would not have an available defence under s 26. The best the defendant could do at trial would be to prove the truth of the two more serious imputations, and rely on that partial justification in mitigation of damages. The plaintiff would likely be entitled to a verdict, an award of (admittedly modest) damages and some or all of their costs.

In short, plaintiffs could circumvent the operation of s 26 as originally formulated and deprive the contextual truth defence of its intended utility by pleading any candidate contextual imputations in their statement of claim, or by amending their statements of claim in order to add, as imputations of which they complain, any contextual imputations identified by a defendant in their defence.

In Kermode, Simpson J thought these implications required urgent correction:

So strongly am I of the view:

The New South Wales Court of Appeal dismissed an appeal from Simpson J’s decision, ⁴⁶ but departed from her view that the implications were unintended. McColl JA, with whom Beazley and Giles JJA agreed, said:

I do not discern any legislative intention … that the s 26 defence was to continue the pleading-back practice which prevailed under s 16 of the 1974 Act. Rather, in my view, the structure of the 2005 Act and the language of s 26 belie any such intention. The New South Wales Attorney General said, in the Second Reading Speech to the Bill that became the 2005 Act, in reference to cl 26 that there would be a defence of contextual truth under the 2005 Act, that there had been one under the 1974 Act and that ‘[t]he purpose of the defence [was] basically to prevent plaintiffs from taking relatively minor imputations out of their context within a substantially true publication’. That position is still open under s 26. ⁴⁷

I have argued elsewhere that the correctness of that statement is open to doubt. In a 2011 article based on the August 2011 submission of the Media and Communications Committee of the Law Council of Australia (of which I was then the national Chair) to the statutory review of the Defamation Act 2005 conducted by the NSW Attorney-General’s Department, I wrote (omitting footnotes):

None of the sources cited by McColl JA indicate a clear view regarding Parliament’s intention in respect of the relevant question. Rather, it seems that no attention was paid by any legislature to the potential implications of the inclusion of the words ‘in addition to’ in s 26(a).

In its December 2019 Background Paper produced for the purposes of the Council of Attorneys-General Review of the Model Defamation Provisions, the Council sided with Simpson J’s view, and recommended that s 26 be amended ‘to ensure it operates as intended, allowing a defendant to “plead back” imputations raised by the plaintiff’. ⁴⁹ The Council noted that such amendments had been strongly supported by all stakeholders who had made submissions on the point. ⁵⁰

On 27 July 2020, the Council approved the Model Defamation Amendment Provisions 2020. Those provisions came into operation in most Australian States and Territories on 1 July 2021.

As amended, s 26 provides:

1. It is a defence to the publication of defamatory matter if the defendant proves that—

a. the matter carried one or more imputations that are substantially true ( contextual imputations ), and

The Explanatory Note accompanying the Defamation Amendment Bill 2020 (NSW) stated that the aim of the amendment to s 26 was ‘to clarify that a defendant may plead back imputations relied on by the plaintiff as well as those relied on by the defendant to establish the defence of contextual truth’. ⁵¹ In his second reading speech on that Bill, NSW Attorney-General, Mark Speakman, said that the Bill replaced the current s 26 ‘with a revised defence of contextual truth that makes it clear that a defendant may plead back imputations raised by the plaintiff’. ⁵²

There is no reason to doubt that the reformulation of s 26 gives effect to that expression of Parliamentary intent. By removing the requirement that contextual imputations be ‘in addition to’ those of which the plaintiff complains, and clarifying that contextual imputations may include such imputations, it will no longer be possible for plaintiffs to circumvent the potential operation of the defence by suing on all imputations carried by a publication, including imputations that are substantially true.

To return again to the example I gave earlier, the reformulated provision means that a defendant who has published true imputations that the plaintiff is a murderer and a thief, but a false imputation that the plaintiff is a jay-walker, will likely have a defence of contextual truth under s 26, irrespective of the imputations of which the plaintiff complains.

As reformulated, however, it seems to me that s 26 goes considerably further than remedying the defect in the original drafting. My contention is that the reformulated defence has also reinstated the Polly Peck principle in Australia and substantially (but not entirely) done away with the confusion arising out of the Hore-Lacy line of authorities.

Those conclusions follow, in my view, from the removal of all qualifiers on the imputations that may serve as contextual imputations for the purposes of the s 26 defence.

Section 26 and the Polly Peck Principle

To qualify as a contextual imputation under s 26(a) in its original form, the imputation had to be ‘in addition to’ an imputation of which the plaintiff complains, and an ‘other’ imputation (i.e. an imputation ‘other’ than one of which the plaintiff complains) that was substantially true.

For some time, there was a degree of uncertainty in the authorities about the extent to which a common sting imputation (i.e. an imputation that could have been relied upon in accordance with the Polly Peck principle) could qualify as a contextual imputation for the purposes of a defence under s 26.

That uncertainty was quelled, at least in New South Wales, in Fairfax Media Publications Pty Ltd v Zeccola (‘Zeccola’). ⁵³ The plaintiffs had (relevantly) sued on three imputations, to the effect that they had dishonoured commitments by not returning profits to producers of Australian films, withheld returns due to producers of successful films, and not distributed a film despite having agreed to do so. The defendant pleaded a contextual imputation to the effect that the plaintiffs had defaulted on payment obligations to producers of Australian films. McCallum J struck-out the defendant’s contextual truth defence, on the basis that the ‘sting is, in my view, not of a different kind — it is essentially the same’. ⁵⁴

That decision was reversed on appeal. McColl JA, with whom Macfarlan JA and Sackville AJA agreed, held that McCallum J had erred. McColl JA said (omitting footnotes):

[83] An imputation of general financial default is capable of being substantially different from the specific instances of such default the respondents have selected. One way of testing that issue is to determine what the party pleading justification or contextual truth would need to prove in order to justify the imputation. That test can be satisfied even if the same evidence proves the truth of each imputation.

The ‘differ in substance’ test has since been described by McCallum J as ‘undemanding’. ⁵⁶

A consequence of that undemanding test is that, in an appropriate case, a defendant can plead, as a contextual imputation for the purposes of a defence under s 26 in its original form, an imputation that could have been pleaded and justified in accordance with the Polly Peck principle.

Recall, again, the facts in Khashoggi. On those facts, it would be open to the defendant to plead a defence under s 26 in its original form, relying on a contextual imputation of general promiscuity in answer to the plaintiff’s complaint of a specific imputation of adultery. If, at trial, the defendant established the substantial truth of the contextual imputation, then the question for the tribunal of fact would be whether, by reason of its substantial truth, the specific imputation of adultery complained of by the plaintiff has further harmed the plaintiff’s reputation. If the tribunal of fact concluded that it did not, then the defendant’s s 26 defence would succeed and the proceeding would be dismissed. If, on the other hand, the tribunal of fact concluded that it did, then the s 26 defence would fail and plaintiff would be entitled to judgement and an award of damages, with the damages being mitigated by the evidence adduced by the defendant in support of the failed s 26 defence.

In effect, therefore, the ‘differ in substance’ test as explained in Zeccola can operate, in a case involving a publication that carries a specific imputation (complained of by the plaintiff) and a more general imputation with a common sting (not complained of by the plaintiff) in the same way as the Polly Peck principle. What was abolished at common law by the post-Chakravarti authorities was restored in part by the adoption of the undemanding ‘differ in substance’ test for contextual imputations in s 26 in its original form.

Because of the drafting deficiency in the original s 26 defence, however, a plaintiff could circumvent its operation by suing on both a (false) specific and a (true) more general imputation with a common sting, thereby depriving the defendant of the ability to rely on the more general imputation as a contextual imputation. In such a case, the defendant would have a defence of justification at common law or under s 25 in respect of the general, but not the specific, imputation; and the plaintiff would be entitled to a verdict in respect of the specific imputation, with damages mitigated, substantially or perhaps entirely, having regard to the substantial truth of the general imputation.

The reformulated s 26 defence removes this possibility of different outcomes based upon the manner in which the plaintiff pleads their case. On a plain reading of the new section, defendants may rely, as contextual imputations, on any imputations carried by the matter complained of, whether exactly the same as, a nuance of, carrying a common sting with, or entirely different from, the imputations of which the plaintiff complains. A contextual imputation need no longer ‘differ in substance’ from an imputation complained of by the plaintiff.

Take, yet again, the facts in Khashoggi. Under s 26 as reformulated, it would be open to a defendant in such a case to plead, as a contextual imputation, an imputation of general promiscuity. Nice questions about whether those imputations ‘differ in substance’ have fallen away by the removal of the former requirement that a contextual imputation be ‘in addition to’ and ‘other’ than an imputation of which the plaintiff complains. Further, because pleading back is now permitted, the operation of the defence can no longer be circumvented by the plaintiff electing to complain of the general imputation, even though it is substantially true.

If the defendant were able to establish the truth of a contextual imputation of general promiscuity at trial, it would be a matter for the tribunal of fact to determine whether, having regard to the substantial truth of that imputation, the imputation of specific adultery has further harmed the reputation of the plaintiff. If the tribunal of fact were to find that it had not, then the defendant would have a complete defence of contextual truth. If the tribunal of fact were to find that it had, then the defendant would be entitled to rely on the truth of the contextual imputation in mitigation of damages. The outcome would be the same, regardless of the decision made by the plaintiff in respect of the imputations pleaded in the statement of claim.

Further, whatever the outcome in the particular case, the result would be indistinguishable, in practice, from the operation of the Polly Peck principle.

Section 26 goes considerably further than the Polly Peck principle, of course: a contextual imputation for the purposes of the statutory defence need not have a common sting with an imputation of which the plaintiff complains; it may be entirely distinct from any imputations of which the plaintiff complains.

Section 26 and the Hore-Lacy Line of Authorities

The ‘differ in substance’ test for determining contextual imputations for the purposes of the defence in s 26 in its original form operated to preclude defendants from relying on contextual imputations that were nuances of an imputation complained of by the plaintiff — that is, contextual imputations of the kind that a defendant could plead and justify in accordance with the Hore-Lacy line of authorities. That is because a nuance of an imputation is necessarily not ‘different in substance’ from the imputation itself. ⁵⁷

A further consequence of the reformulation of the s 26 defence seems to me to be that it has substantially, though not entirely, reversed that position as well as, in many cases, at least for practical purposes, the effect of the Hore-Lacy line of authorities.

Under the reformulated defence, a defendant may rely as a contextual imputation on any imputation carried by the publication, including an imputation complained of by the plaintiff. It must follow that a nuance of an imputation complained of by the plaintiff may also be relied upon as a contextual imputation for the purposes of the reformulated defence.

Provided that a contextual truth defence survives any strike-out application, evidence will be able to be properly admitted at trial in support of each contextual imputation relied upon by the defendant. If the defendant proves that, having regard to the substantial truth of the contextual imputations, any residual imputations of which the plaintiff complains do not further harm the reputation of the plaintiff, then the defendant will have a complete defence. If the defence fails, because the tribunal of fact finds that the residual imputations have further harmed the reputation of the plaintiff, then the plaintiff will be entitled to a verdict and damages, but the substantial truth of the contextual imputations will be able to be relied upon by the defendant in mitigation.

Assume, by way of example, that a publication is capable of carrying both guilt and reasonable grounds imputations in respect of at least two distinct charges — say of murder and jay-walking. In such a case there are four possibilities: that the publication conveys that the plaintiff is a murderer and a jay-walker; that the plaintiff is a murderer and there are reasonable grounds to believe that the plaintiff is a jay-walker; that the plaintiff is a jay-walker and there are reasonable grounds to believe that the plaintiff is a murderer or that there are reasonable grounds to believe that the plaintiff is both a murderer and a jay-walker.

If the plaintiff were to sue only on guilt imputations — that is, imputations of murder and theft, it would be open to the defendant to plead, as contextual imputations, that the publication carried contextual imputations that there are reasonable grounds to suspect that the plaintiff is a murderer and a jay-walker. Under the defence as reformulated, it is of no moment whether those imputations differ in substance from or are mere nuances of the imputations complained of by the plaintiff.

Such a pleading might attract a strike-out application, on the premise that the defence is bound to fail. Such an application, however, would likely be determined by assuming that, at trial, the publication is found to carry only the least serious of the imputations complained of by the plaintiff and that imputation is not proved to be true, and all of the contextual imputations pleaded by the defendant and that those imputations are each proved to be true. ⁵⁸

On the facts posited, the court would thus assume, on a strike-out application, that the publication falsely imputed that the plaintiff is a jay-walker, but correctly imputed that there are reasonable grounds to believe that the plaintiff is a murderer and a jay-walker.

It is by no means clear that such a strike-out application would succeed: the court would have to be satisfied, as a matter of law, that a tribunal of fact at trial, having heard all of the facts and circumstances, ⁵⁹ would be bound to conclude that the plaintiff’s reputation had been further harmed by the publication of the jay-walking imputation, despite the substantial truth of the imputation that there were reasonable grounds to believe that the plaintiff was both a murderer and a jay-walker.

Assuming the defence survived a strike-out application and went to trial, evidence could be adduced in support of the substantial truth of the contextual imputations. The defence would succeed if the tribunal of fact concluded that, because of the substantial truth of the contextual imputations, the residual imputations of which the plaintiff complains and which were found to be conveyed have not further harmed the plaintiff’s reputation. The defence would fail if the tribunal of fact concluded that the residual imputations have further harmed the plaintiff’s reputation, despite the substantial truth of the contextual imputations, but in that event, the substantial truth of the contextual imputations could be relied upon by the defendant in mitigation of damages.

The Hore-Lacy line of authorities would remain relevant, however, in cases where a s 26 defence would not survive a strike-out, such as cases involving publications that are capable of imputing guilt and reasonable grounds imputations in respect of a single charge. A defendant could not rely, in answer to a complaint that a publication imputed murder, on a contextual imputation to the effect that the publication imputed that there were reasonable grounds to believe that the plaintiff was a murderer. In such a case, a defence under s 26 would be bound to fail, because an imputation that a plaintiff is a murderer would necessarily further harm the reputation of the plaintiff, even if it be true that there were reasonable grounds to believe that the plaintiff was a murderer. In such cases, the Hore-Lacy line of authorities still has work to do.

These implications of the reformulated s 26 defence of contextual truth — which I think are not yet well understood — while complicated and yet to be worked out in the authorities are, in my view, to be welcomed. By permitting a defendant, other than in cases attracting a successful strike-out application, to plead as contextual imputations any meaning carried by a publication, and then to adduce evidence in support of those meanings at trial, the reformulated defence focuses on what should be in issue in a defamation action where truth is in play: the meaning of the publication taken as a whole, and whether (and if so, to what extent) the publication taken as a whole has damaged the plaintiff’s reputation having regard to the extent to which it is a matter of substantial truth.