What happened
In November 2010 Western Australian police, acting on intelligence, arranged for New South Wales officers to intercept a cardboard box consigned from Sydney that contained two padlocked toolboxes holding 4.981 kg of 84% pure methylamphetamine. The consignment note directed the freight company to call a mobile number ending in 731 to arrange collection. Police substituted rock salt for the drug, reconstructed the box with a listening device inside, and allowed delivery to proceed. Matthew Lothian collected the box in Perth on 16 November after his car ran out of fuel en route. He took it to premises in Falstaff Crescent, Spearwood. Approximately four minutes later the appellant, Pouyan Kalbasi, arrived by bicycle. The listening device captured sounds of the box and toolboxes being opened, locks being cut, and conversation in which Lothian recounted his fuel problems after first texting the appellant. Kalbasi asked for a pipe, sampled the substance, told Lothian "Don't move. I'll come back", and left. He rode into a nearby park, appeared to make a call, and then fled when approached by unmarked police who identified themselves. He was apprehended after a short chase. Police immediately searched the Spearwood premises. They found the opened packaging, nine bags of rock salt in a cupboard, scales, mixing bowls, MSM (a cutting agent), disposable gloves (one pair bearing Kalbasi's DNA inside and out), and other paraphernalia consistent with preparation for cutting and distribution. Kalbasi's DNA was not on other gloves linked to Lothian and his girlfriend Venetia Tilbrook. Travel records, phone data and the 731 number linked Kalbasi to the consignment. He gave no evidence at trial.
Kalbasi was indicted for attempting to possess a prohibited drug with intent to sell or supply contrary to ss 6(1)(a) and 33(1) of the Misuse of Drugs Act 1981 (WA). This was a retrial after an earlier conviction was set aside. At both the trial and retrial all parties, including experienced senior and junior counsel for the defence, laboured under the mistaken belief that s 11 of the MDA (which deems possession of more than two grams of methylamphetamine to be possession with intent to sell or supply unless the contrary is proved) applied to an attempt charge. The trial judge provided a jury aide listing the four elements, including intent. After the close of the prosecution case, defence counsel expressly agreed that if the jury found possession there was "no issue" on the fourth element. The judge directed the jury that proof of possession of more than two grams triggered the presumption, that there was "no issue in this trial about the fourth element", and that they could "give it a tick" without further deliberation. The sole issue fought at trial was whether Kalbasi exercised control or dominion over the entire consignment while present for approximately 37 minutes. Defence counsel's opening and closing speeches, and the so-called "Nissan submission", emphasised that mere presence or knowledge was insufficient and that the prosecution had not proved control. No positive case was advanced that Kalbasi was there only to sample a small quantity for personal use. He was convicted.
On appeal to the Court of Appeal (McLure P, Mazza and Mitchell JJA) the State conceded the direction was wrong in law following Krakouer v The Queen (1998) 195 CLR 202, which held s 11 inapplicable to attempts. The Court of Appeal nevertheless dismissed the appeal under s 30(4) of the Criminal Appeals Act 2004 (WA), the local form of the common-form proviso. McLure P held that once the jury found possession of nearly 5 kg of high-purity drug, conviction with the requisite intent was inevitable. Mazza and Mitchell JJA distinguished errors of "process" and "outcome", found the misdirection was not a fundamental process error, and concluded that guilt was proved beyond reasonable doubt on the whole record.
The High Court granted special leave limited to the proviso question. By majority (Kiefel CJ, Bell, Keane and Gordon JJ) the Court dismissed the appeal, holding that the misdirection did not occasion a substantial miscarriage of justice. Gageler, Nettle and Edelman JJ dissented in separate reasons, each concluding that the removal of the intent element from the jury's consideration constituted a fundamental defect that prevented the appellate court from being satisfied that conviction was inevitable or that the jury's verdict might not have differed.
Why the court decided this way
The majority began by reaffirming Weiss v The Queen (2005) 224 CLR 300 as the governing authority. They rejected the appellant's invitation to qualify or overrule Weiss or to return to the pre-Weiss "lost chance of acquittal" test. The statutory text of s 30(3) and (4) of the Criminal Appeals Act requires the Court of Appeal to allow an appeal if there has been a wrong decision on a question of law or a miscarriage of justice, but permits dismissal if no substantial miscarriage has occurred. Weiss resolves the apparent tension by committing the assessment to the appellate court itself, which must decide whether, notwithstanding the error, the accused's guilt was proved beyond reasonable doubt on the admissible evidence actually led at the trial that was had. The appellate court does not ask what a hypothetical jury would have done. Concepts of "inevitability" or "lost chance" are useful only as pointers to the high criminal standard and the natural limitations of proceeding on the record; they are not alternative tests.
Crucially, Weiss requires consideration of the nature and effect of the error in every case. Some errors (contested credibility cases, failure to leave a defence, or misdirection on a live element) may prevent the appellate court from being able to reach the requisite satisfaction. The majority held that the present misdirection was not of that character. The trial had been conducted on the explicit footing that the only dispute was possession. Defence counsel had conceded that if possession in the legal sense was proved, the inference of intent given the volume would follow, and had agreed that the jury need not concern itself with element four. The "Nissan submission" was directed only to control, not to a limited-purpose sampling hypothesis. There was no evidence raising the possibility that Kalbasi possessed only a small quantity for personal use. The prosecution case, the listening device evidence, the DNA on the inside of the gloves, the flight, and the commercial quantity and purity all pointed to joint control of the entire consignment for onward distribution.
The majority distinguished Krakouer, where the misdirection had reversed the onus on intent in a conspiracy count and intent had been live. They also distinguished Handlen v The Queen (2011) 245 CLR 282 (trial conducted on a legal basis that did not exist) and Quartermaine v The Queen (1980) 143 CLR 595 (failure to direct on an essential element that the jury had not been asked to consider). Here the jury was correctly instructed on all four elements but told, erroneously, that the fourth was proved if the third was. Because the fourth was not live and the facts once possession was accepted compelled the intent conclusion, the error did not deprive Kalbasi of a chance of acquittal that was fairly open. The Court of Appeal had been entitled to infer from the verdict, the way the case was run, and the record that the jury must have found possession of the whole consignment. From that finding intent followed inevitably. Accordingly no substantial miscarriage of justice had occurred and the proviso was properly applied.
The dissenters reached the opposite conclusion by emphasising that the jury had been positively told to "give it a tick" on an essential element and that the broad directions on possession (including temporary and limited-purpose possession illustrated by the library book example) left open pathways to a possession finding that were inconsistent with the intent required for the charged offence. They viewed the removal of the element as a fundamental defect amounting to Kalbasi not being tried for the offence charged.
Before and after state of the law
Before Kalbasi the law was settled in two respects but uncertain in application. Krakouer had clearly held that s 11 of the MDA has no application to attempt charges; the presumption cannot deem intent on an attempt to possess. Weiss had authoritatively reframed the proviso inquiry away from hypothetical jury predictions toward the appellate court's own satisfaction of guilt on the actual record, while insisting that the nature of the error must always be considered. Intermediate courts, however, had begun to treat the Weiss "negative proposition" (that the proviso cannot be applied unless the appellate court is satisfied of guilt beyond reasonable doubt) as if it were both necessary and sufficient, and had formalised a binary "process" versus "outcome" taxonomy derived from Hughes v The State of Western Australia [2015] WASCA 155. That taxonomy risked elevating the negative proposition into a complete test and marginalising the statutory question.
Kalbasi confirms Weiss without qualification and clarifies its application in the specific context of misdirection on a non-live element. The majority expressly disapproved treating "process" and "outcome" as rigid compartments that allow an appellate court to leap from a finding of proved guilt to dismissal without considering whether the error deprived the court of the ability to make that finding justly. The decision reinforces that the conduct of the trial, including forensic concessions and the issues actually litigated, forms part of the "whole record" against which the proviso is applied. It also confirms that a misdirection reversing or removing an onus on a non-live issue will not always preclude the proviso, provided the appellate court can be satisfied that the element was proved beyond reasonable doubt by the facts necessarily found by the jury.
After Kalbasi appellate courts must still begin any proviso analysis by identifying the precise error and its capacity to affect the jury's consideration of live issues. The decision has not altered the underlying Weiss framework but has narrowed the circumstances in which an error on an element will automatically be characterised as fundamental. It has also underscored the importance of the trial record as it actually unfolded rather than as it might have unfolded had counsel appreciated the correct legal position.
Key passages with plain-English translation
The joint judgment contains several passages that encapsulate the reasoning. One states: "Weiss requires the appellate court to consider the nature and effect of the error in every case. This is because some errors will prevent the appellate court from being able to assess whether guilt was proved to the criminal standard." In plain English, the High Court is saying that judges cannot simply declare "the evidence looks strong so the mistake does not matter"; they must always examine exactly what went wrong and whether that wrong prevents them from fairly deciding guilt themselves on the existing record.
Another key passage reads: "The sole issue in the way the trial was run was proof that the appellant was in possession of, in that he was exercising control (by himself or with Lothian and, perhaps, Tilbrook) over, the substitute 'drugs' in the cardboard box. The Court of Appeal was correct to reason that proof beyond reasonable doubt that the appellant attempted to possess nearly 5 kg of 84% pure methylamphetamine compelled the conclusion that it was his intention to sell or supply it to another." Translation: because everyone at trial treated intent as automatic once possession of that enormous, high-purity haul was proved, and because the defence never suggested Kalbasi was only there for a personal sample, the appellate court could safely conclude that the jury's possession finding carried the intent finding with it. The mistake therefore changed nothing of substance.
A further passage addresses the forensic concession: "Senior counsel's evidence is that he would not have consented to a direction that intention was not in issue had he appreciated that s 11 of the MDA did not apply... Senior counsel did not say that he would have conducted the appellant's case differently in any other respect... McLure P, rightly, concluded that against this background the inference to be drawn is that a considered and justifiable forensic decision was made not to run a case that the appellant's conduct was consistent with an attempt to possess a small quantity of the drug for his own use." Plain English: even though counsel later said he would not have agreed to the direction, he never suggested he would have run a different positive case. The trial judge and Court of Appeal were entitled to treat the concession as a deliberate tactical choice, not an accident that vitiated the whole trial.
These passages illustrate the majority's insistence on fidelity to the actual trial that occurred rather than a hypothetical one that might have been run with perfect legal knowledge.
What fact patterns trigger this precedent
Kalbasi is triggered when three conditions coincide. First, there must be a misdirection or wrongful removal of an element of liability, commonly by erroneous application of a statutory presumption or reversal of onus. Second, that element must not have been "live" at trial: the parties must have conducted the case on the explicit or implicit understanding that proof of one element resolved the other, typically evidenced by concessions in open court, jury aides, or closing addresses that narrow the dispute to a single factual contest (here, control). Third, the appellate court must be able to conclude on the whole record, including the verdict itself and the circumstantial strength of the case, that the facts necessarily found by the jury on the live issue compel the conclusion on the misdirected issue beyond reasonable doubt.
Typical triggers therefore include large-scale commercial drug importations or possessions where the quantity alone makes innocent intent implausible, cases in which defence counsel has positively conceded an element for tactical reasons, and retrials where the defence has abandoned lines run at an earlier trial (as Kalbasi had given evidence of personal-use sampling at his first trial but not the second). The precedent does not apply where the misdirected element was actively contested, where credibility is central, where a partial defence was not left to the jury, or where the record leaves open a rational pathway to acquittal consistent with the erroneous direction. It is particularly inapplicable where the trial has been run on a wholly wrong legal basis (Handlen) or where the jury has never been asked to consider the real offence charged (Quartermaine).
How later courts have treated it
Subsequent Western Australian Court of Appeal decisions have treated Kalbasi as confirming and refining rather than revolutionising Weiss. In The State of Western Australia v Rayney (No 3) [2019] WASCA 78 the Court cited Kalbasi for the proposition that the proviso analysis must always begin with the precise nature of the error and its capacity to affect the jury's consideration of live issues; a misdirection on a non-contentious element was held not to produce a substantial miscarriage where the remaining issues had been fairly tried. In drug cases such as R v Nguyen [2020] WASCA 79 the Court has repeatedly invoked the Kalbasi emphasis on the conduct of the actual trial, refusing to speculate about unadvanced "sample for personal use" hypotheses where they were not put to the jury.
Interstate courts have also embraced the decision. Victorian and New South Wales intermediate appellate courts have cited Kalbasi alongside Weiss and Filippou v The Queen (2015) 256 CLR 47 for the proposition that the negative proposition is not a sufficient condition and that rigid process/outcome taxonomy should be avoided. In Bakar v The Queen [2021] NSWCCA 212 the New South Wales Court of Criminal Appeal distinguished Kalbasi on the basis that the intent element there had been live and the quantity of drug far smaller, so that a sampling hypothesis remained rationally open. Federal Court appeals involving Commonwealth drug offences have used Kalbasi to emphasise that forensic concessions by experienced counsel carry weight in the proviso analysis.
Overall, Kalbasi has been followed rather than distinguished in the majority of reported decisions. It has not been criticised or confined. Later courts treat it as the leading modern High Court authority on the application of the proviso to misdirections on non-live elements in circumstantial cases.
Still-open questions
Several questions remain unresolved after Kalbasi. First, how far counsel's post-trial affidavits explaining forensic decisions can be taken into account remains unclear. The majority noted the desirability of receiving such evidence had not been debated but treated the affidavit as unchallenged; however, they did not lay down any general rule about when such evidence is admissible or what weight it bears when it seeks to rewrite the trial.
Second, the precise boundary between a "non-live" element and one that is merely not the subject of positive contradictory evidence is unsettled. The majority accepted that a trial judge is not required to direct on an element that is not in issue, yet the dissenters emphasised that counsel cannot concede a matter of law to the accused's disadvantage. Future cases will need to determine when a concession induced by mutual legal error crosses into the Quartermaine category of the jury never having been asked to decide the real offence.
Third, the interaction between Kalbasi and jury directions on "extended" or "temporary" possession remains open. The dissenters placed heavy reliance on the library-book analogy; the majority held it was not apt to mislead on the facts. In future cases with less overwhelming quantities or different circumstantial evidence, the breadth of possession directions may again become decisive.
Finally, the continuing utility of the "process" and "outcome" language endorsed in Hughes but criticised in Kalbasi has not been definitively settled. While the majority accepted such classifications "may or may not be helpful", intermediate courts continue to use them. Whether a future High Court will expressly disapprove the Hughes taxonomy remains to be seen. Until then, practitioners must treat Kalbasi as a reminder that the statutory question is always whether there has been a substantial miscarriage of justice, judged by reference to the trial that actually occurred.