Anyone tempted to believe George Pell did what he was convicted of doing should read first the majority judgment of the Court of Appeal majority (“Judgment”), next the fuller transcript of the complainant’s allegations that is given in paras. 415-55 of the dissenting judgment (“Dissent”), and then the Wikipedia account (with numerous links) of Operation Midland.
If you take this short tour, you will see the Judgment fall apart under your eyes. The Judgment’s sequencing (Falsity, Improbability, Impossibility) reverses the rational order of treatment. Its handling of Archbishop Pell’s alibi defence concludes abruptly in para. 143 by placing the onus of proof exactly where the law quoted in para. 142 says it cannot be: on the defence. Its construction of a five or six minute window of opportunity for the Archbishop to commit singularly vile offences against two thirteen-year-old boys, in the Priests’ Sacristy, has a similar incoherence thinly veiled behind an “of course” and an evasive “taking the evidence as a whole”.
A brief account of those three ways the 352-paragraph Judgment goes wrong will indicate how the jury’s one-word verdicts could be as wrong as one should conclude they were.
Of course, there is another secure route to that conclusion: read the Dissent. It brings to light many other reasons to reject the complainant’s allegations. But it is long and winding. Here, then, is one shorter route.
One: Rational sequencing reversed.
The Judgment went wrong by considering first the defence’s contention that the complainant’s tales of rapes and other assaults by the Archbishop of Melbourne were false, along with the defence’s alternative explanations of that falsity: dishonest fabrication or honest fantasy (or some combination of these). The defence had no obligation to suggest, and did not begin to suggest, any motive for fabricating or any cause for fantasising. On the defence case, the falsity of the allegations is a conclusion from all the evidence taken together: that is, from the gaps and alterations in the allegations, from their inherent improbability, and from their incompatibility with the wealth of evidence that the Archbishop was absent from the sacristies at the relevant times (impossibility) and that, in all probability, numerous other people were coming and going and/or unrobing and/or sitting about in the Priests’ Sacristy at those times (impossibility or improbability).
By treating falsity as a distinct argument (rather than a conclusion from other arguments) – and also by treating it before improbability and impossibility – the Judgment displays deep confusion about the case’s basic logic, aborts its own rational consideration of the defence, and effectively reverses the onus of proof. The defence had presented  the three matters in a rational and cumulatively inter-connected sequence: (A) the testimony’s improbability both inherently and as given with inconsistencies, opportunistic embellishments, and sheer mistakes, (B) its impossibility as demonstrated by much counter-evidence, and (C) the appropriate conclusions: the testimony, however “compelling” as delivered , , , is certainly false, or most probably false, and, at any rate, the possibility of its falsity is so real that the jury should have doubted it, and Pell in both law and justice should have been acquitted. But the Judgment , in reversing the sequence, also practically eliminated the inter-connections and cumulation, that is, the rational bearing of (B) on (A), and of (A) and (B) together on (C). And it made this reversal for no stated reason, but just as something “it is convenient” to do .
Under “falsity”, thus misconceived, the Judgment dealt with a knot of issues that could indeed be rightly considered – but only provisionally considered – before considering the evidence about the Archbishop’s absence (impossibility) and other people’s presence (improbability or impossibility). This is the knot of issues about the internal (im)plausibility and (in)coherence of the complainant’s testimony and the (in)consistency of his several iterations of it. Watching twice (like the jury) the video of two of those iterations, the Judgment’s authors found him credible and true.
The phrase is not theirs but sums up the conclusions reached and opinions conveyed in this part of the Judgment. “Credible and true” is the phrase actually used by the very senior officer of the Metropolitan Police’s Operation Midland to describe the detailed testimony given to multiple police officers on multiple occasions by “Nick” (one Carl Beech), presenting himself as a victim/survivor and/or witness of sexual abuse, and sex-murders, witnessed by him from 1975 (aged 7) to 1984 (aged 16), at the hands of a former prime minister, former Heads of MI5 and MI6, a former Home Secretary, a former Chief of the Defence Staff, and other named persons of similar standing in British public life. In early 2015, the same year the Victorian complainant came forward to testify to Victorian Police that he was a victim/survivor of Archbishop Pell, those of the British persons just mentioned still living had their lives irreparably damaged by 20-hour police searches and public police accusations all made in total reliance on “Nick’s” testimony. And in 2019, while the Court of Appeal was hearing Cardinal Pell’s appeal and writing the Judgment, Carl Beech, long called by the police and media a “victim/survivor”, was being tried (over ten weeks) and in due course sentenced to 18 years imprisonment for perverting the course of justice. For despite Beech’s ability to describe places where these public figures were likely to have been, there turned out to be no truth in his accusations – though, while sufficiently insulated from counter-evidence and accurate contextualisation, they had been judged by experienced police detectives to be truthful and “credible and true”.
The Pell Judgment declares its authors’ entire satisfaction with the truthfulness and accuracy of the complainant, and it does so before turning to confront any of the contextualising counter-evidence. The effect, despite its routine preliminary affirmations that the defence has no onus  (also ), is clear: to place on the defence the burden of proving the testimony false, the onus, that is to say, of (in one of the Judgment’s several erroneous formulations) “establishing the certainty which the [defence’s] argument of impossibility asserted” . That indefensible reversal is the topic of sec. two below.
Meanwhile, the Judgment finds that the credibility of the complainant’s testimony was “considerably enhanced by the accuracy of his description” of the sacristy in which he said he was raped etc., and by the fact, “more striking still”, that because of redecorating works, the Archbishop was, unusually, having to robe and disrobe in that sacristy – the Priests’ Sacristy – rather than in the adjoining one reserved to his use . These two facts were :
independent confirmation of A’s account of having been in the Priests’ Sacristy in that period. There was nothing to suggest that his knowledge of those matters could have been obtained otherwise. A’s evidence was that he had never been in the Priests’ Sacristy before.
These bits of “independent confirmation” (what used to be called corroboration) each fall apart.
As to the first: the Judgment promptly contradicts both its own “nothing to suggest” claim and its claim about “A’s evidence”. For under cross examination, A accepted  (, , -) that he had been given a cathedral tour, on becoming a choir boy, and accepted that such a tour would have included (though he said he could not recollect this) a visit to the Priests’ Sacristy.
As to the second fact: the complainant’s testimony in no way suggested that Pell had entered to disrobe; it just said  () he entered, “planted himself in the doorway” (the doors of the sacristy from the corridor leading back to the sanctuary, aisles and nave) and challenged the boys. Moreover, there was a period of months in which the Archbishop was obliged to use the Priests’ Sacristy for robing and disrobing , and nothing to suggest that at other times he was never to be seen in the Priests’ Sacristy heading to or from his own adjacent sacristy (further from the cathedral’s liturgical activity) via the door in the partition-wall between the sacristies, or conversing in either sacristy with priests or altar servers . And the complainant as a choirboy must have gone right past the main door of the Priests’ Sacristy (not to mention the main door of the Archbishop’s Sacristy) on many occasions, at times when it was likely to be open before or after services.
The reluctantly admitted preliminary choir tour weakens to the point of extinguishing the corroboration which the Judgment finds in the complainant’s knowledge (such as it was) of sacristy layout. But even setting aside the tour, nothing suggested that – at any time while he was a choirboy – he might not have peeked or ducked into the Priests’ Sacristy and seen its arrangement, in an escapade, perhaps of seconds, perhaps even of minutes, perhaps accompanied or alone, just conceivably even for wine-swigging, an escapade that included no confrontation with any Archbishop (or with anyone) and no oral rape. The “independent confirmation” gets nowhere near tending to confirm any claims concerning the Archbishop.
About those claims, and the question of “independent support” for them, the Dissent rightly summarises the position:
“There was no forensic, or other objective evidence, to support [the complainant’s] account”; indeed, “the jury were invited to accept his evidence without there being any independent evidence to support it” ,”…entirely unsupported…”, “no supporting evidence of any kind” , , .
And the Judgment, while clutching at straws to find confirmation, passingly admits the falsity of the complainant’s denials that he had ever been in the Priests’ Sacristy before or after the day on which, he said, he and another boy (now dead) swigged wine and were raped there. The Judgment reports and ignores this admission as blandly as it ignores the evidence - that his testimony about the colour of the wine and of its bottle was all untrue, and that both his description and his recollection of the relevant part of the sacristy, so far from being impressively accurate, were quite inaccurate -.
Two: Onus of proof reversed
The shortest of all routes to discovering that the Judgment has gone catastrophically wrong is to read paras. 139 to 143 and para. 151. Para. 139 summarises one of the ways in which the defence argued that the alleged offending in the Priests’ Sacristy was impossible: Archbishop Pell was at the relevant time far away at the west door with his master of ceremonies, Fr. Portelli, meeting and greeting worshippers. Portelli’s testimony (not to mention the testimony of many other witnesses) was cogent evidence of that, and if true constituted an alibi for Cardinal Pell.
In para. 140, the Judgment remarks that the concepts of alibi, impossibility and (lack of) opportunity are “of course, closely inter-connected.” But it goes on to say that the defence at the trial had (at least in its closing) avoided the word “alibi”, had asked the trial judge not to use it, and  had not asked for a direction to the jury in the form appropriate to alibi defences. Without making any comment on those features of the conduct of the defence at the trial, or on the fact that the prosecution in its final trial address  had called some of the evidence “alibi”, the Judgment then and there  sets out the law applicable to alibi defences. Neither here nor anywhere else did the Judgment suggest that the defence of George Pell is disqualified from relying on this law.
To state it, the Judgment uses a source different from the Dissent’s sources [396, , ,  but with precisely the same legal content. The jury cannot rightly convict unless the prosecution has “remove[d] or eliminate[d] any reasonable possibility” that the accused was not at the alleged crime-scene (the sacristy) but somewhere else instead (the west door).
Then, with a startling lurch, the Judgment goes straight from stating that law to stating  its own position, essentially its fundamental conclusion about the whole case:
Having read all of the opportunity evidence and watched some of it, we are not persuaded that the evidence of any individual witness, or the evidence taken as a whole, established impossibility in the sense contended for by the defence.
The next sentence adds: “In pt II of the reasons, we explain that conclusion by reference to the evidence relied on in support of each of the individual impossibility contentions.” To wrap up part I, its main part, the Judgment proceeds to give an example - of its way of dealing with an “individual impossibility contention”, and then circles back to the general significance of the impossibility v. possibility argument. The substance of para. 143’s astonishing transfer of the burden of proof to the defence is now repeated :
As we have said, the onus of proof required the prosecution to defeat [the argument of impossibility]. It was both necessary and sufficient for that purpose to persuade the jury that the events were not impossible and that there was a realistic opportunity for the offending to occur.
Finally, in relation to “opportunity” (the remaining facet of the alibi – impossibility – no opportunity complex), para. 170 repeats that the prosecution need do no more than establish a “realistic opportunity”. What had emerged, says para. 170, was “not a catalogue of ‘impossibilities’…but…of uncertainties and possibilities…. Plainly enough, uncertainty multiplied upon uncertainty does not – cannot – demonstrate impossibility.”
Now it is significant that some of these “uncertainties” were rustled up out of witnesses’ syntax and, like other “uncertainties”, were in any syntax and on any view unchallenged near-certainties. But that is not the subject of this article. Here the point is that – as is laid down in the legal rule, quoted in para. 142 but then left hanging enigmatically in the air – it was not for the defence to “demonstrate” or “establish” impossibility. Nor was it sufficient for the prosecution to establish possibility in the strict sense of “not impossible”, or even to establish realistic possibility in the sense of “realistically, or in reality, not impossible”. There is a wide chasm between, on the one hand, the Judgment’s there was a realistic possibility that the rapes could have happened and, on the other hand, the law’s standard, quoted without demur by the Judgment, a standard which demands a finding of not guilty unless there was no realistic possibility that he was away from the sacristy and additionally no realistic possibility that at least one other person (concelebrating priest, altar server, sacristan) was in the sacristy or (parishioner) at its open door for even a moment in the five or six minutes after Mass on 15 December 1996.
Three: Evidence wrongly assessed
Still, it would be grossly mistaken to think that this defendant was entitled to be acquitted only, or even mainly, because of some legal rule about alibi defences, a rule perhaps surprisingly demanding on the prosecution. The point of the preceding paragraphs was to show, as briefly as possible, how very unsatisfactory the Judgment is in discharging its primary responsibility to apply the law coherently to the case before the court.
To see why everyone should think that George Pell not only was legally entitled to be acquitted but simply did not do any of the criminal acts alleged, one may, once again, take the long route of reading the Dissent. Its conclusions are expressed with great restraint -; in substance: anyone reasonably considering the evidence should doubt – reasonably doubt – his guilt. Very illuminating is the abundant evidence the Dissent assembles, and the report it gives  of the impression made upon this careful and experienced judge by watching on video many witnesses – both the complainant and a selected eleven of the many witnesses to practical impossibility, alibi and lack of opportunity. (Tellingly, the Judgment alludes to its authors’ impressions on watching the complainant on video, but about other witnesses watched is silent.)
But besides that long Dissenting route, there is, again, a short traverse: a brief examination of a passage in which, not the Dissent, but the Judgment is handling the facts, just one of the many sets of facts on which it touches.
This particular passage - concerns the question whether there really could have been a period of five or six minutes, right after the conclusion of Mass, in which this archbishop could be alone in the Priests’ Sacristy, having his way first with B (now deceased, having denied ever being interfered with), then with A, and then in another sordid way with A, uninterrupted by anyone (and after instructing A to “undo his [A’s] pants and take them right off”  so that the Archbishop could commit this third set of offences – and be yet more irretrievably exposed as a wicked criminal if anyone came in or even glanced in). To find the five or six minutes that, according to A, the offences against him and B took to complete, the Judgment deploys an “of course”.
Now the Judgment had earlier used an “of course” to somewhat similar effect :
It is, of course, of the very nature of an impossibility argument that it seeks to establish with sufficient certainty that the events could not have happened as alleged…
That “of course” had the effect of muffling what was going on (as was shown in the last section): shifting to the defence the onus of “establishing [alibi/impossibility/lack of opportunity] with sufficient certainty”, and in the process relieving the prosecution of its burden of disproving alibi, etc.
So too, with the “of course” in para. 296 (here quoted in full, with emphases, exclamations and interpolations added):
The effect of the servers’ evidence was that the unlocking of the Sacristy door, and their bowing to the Crucifix [inside that Sacristy, to mark the end of their procession duties and the beginning of their altar duties], occurred soon after the procession [to the west door and then back, whether inside or outside, to the east end] finished and that, by the time they returned [from the Sacristy!] to the sanctuary to assist [the sacristan] Potter [in clearing the sanctuary], the door was already unlocked. [!] On that view, it was quite possible for the Sacristy to have been unlocked and unattended at around the time A said he and B broke away from the procession. The clearing of the sanctuary had, of course, to await the end of the private prayer for [= of] parishioners. The Crown case as presented to the jury was that ‘there is this hiatus, this gap’ during which the first incident [oral rapes etc. over 5 to 6 minutes] had occurred.
And the Judgment committed itself to that Crown case :
..taking the evidence as a whole, it was open to the jury to find that the assaults took place in the 5-6 minutes of private prayer time, and that this was before the ‘hive of activity’ described by the other [= other than Potter] witnesses began.
In framing its theory of guilt in this way, the Judgment was selecting one of two different accounts given by Potter. (He was testifying when aged about 84, about events over 20 years earlier, in 1996, when he had already been sacristan for nearly 35 years.). The Judgment silently ignores one of the accounts and relies on the other. But each is incompatible with the complainant’s story.
On one account, reported by the Dissent in para. 504, Potter unlocked the Priests’ Sacristy door almost as soon as the procession (with the choirboys including A and B near the front, followed by six to twelve altar servers, any priests, and at the end the Archbishop) set out from the sanctuary area, moving down the central nave to reach the west door. Other witnesses supported this timing explicitly  or implicitly , . But it was not deployed by the prosecution or the Judgment. For although it entailed that the Sacristy was unlocked and might therefore, as a matter of physical possibility, be entered (as alleged by the complainant A) by errant choirboys and an even more errant Archbishop arriving on the scene after the procession had concluded or nearly concluded, it equally entailed that, at that same (alleged) time and at all times compatible with the accusations, the Sacristy was in constant use, first by the sacristan and then by both him and some or all of the altar servers.
So, instead, the prosecution and the Judgment rely upon sacristan Potter’s other account. Summarised by the Dissent in para. 505, it is given by the Judgment in para. 293 like this:
It was common ground that Potter was the person who unlocked the Priests’ Sacristy [within which the alleged rapes etc. occurred on 15 December 1996] and that he did so after Mass. His evidence was that, after the choir and clergy had processed to the west door, he would go the sanctuary, where he would wait until parishioners had finished what he called their ‘private time’ for prayer after the service. This was typically a period of five or six minutes. He would then take books from the sanctuary and unlock the door to the Priests’ Sacristy. He would then return to the sanctuary to gather up the sacred vessels and – sometimes with the assistance of the altar servers – would take them back to the Priests’ Sacristy.
Thus the prosecution’s and the Judgment’s theory rests entirely on (i) accepting, “of course”, Potter’s somewhat disputed evidence that sanctuary clearing (and the resultant traffic to and from the Priests’ Sacristy) was delayed for five or six minutes after the end of Mass (to permit parishioners’ private prayer), while simultaneously (ii) overlooking the necessary implications of another integral part of his same account, a part disputed by no other witness and stated without a qualm in para. 293, as we have just seen: during that “5-6 minutes hiatus for parishioners’ prayer” the Priests’ Sacristy door was locked.
So, part by part or “taken as a whole”, the evidence as summarised in the Judgment left no room at all for the Judgment’s conclusion in . Neither of Potter’s accounts left time – still less the five or six minutes of solitude alleged by the complainant – for someone to commit offences against choirboys in the Priests’ Sacristy. Potter’s “5-6 minutes of prayer time” account affirmed, without challenge, that the Sacristy door was locked until the end of that time, and the only other evidence (Potter’s and others’) about the “hive of activity” in that Sacristy was that it began much sooner than “5-6 minutes” after Mass.
The Judgment’s conclusion, it is worth adding, is excluded also, and equally completely, by a plain fact unconsidered in the Judgment but obvious from all the evidence and the Judgment’s map of the cathedral. “A” said that his escapade with “B” began when the procession had nearly ended. But by that time, the “5-6 minutes” prayer time allotted by the Judgment for the assaults had been used up.
The Judgment’s contention that the complainant’s evidence was not false should only have been made by reference to the whole of the evidence, and not just by reference to his appearing credible. The contemporaneous case of Beech simply illustrates the point: accusations made by a complainant about sexual abuse who was very credible to many experienced officers were shown to have been entirely false.
Despite their recitations of the rules about onus, the majority shifted the onus onto the defence by saying that he had failed to establish that certain matters were improbable or (practically) impossible.
In satisfying themselves that there was a five- or six-minute window of opportunity – an unlocked and traffic-free door – the majority deployed a reference in the sacristan’s evidence to “five or six minutes”. They not only ignored other evidence from him and other witnesses that ferrying of stuff from sanctuary to sacristy would begin immediately the procession left the sanctuary – but also failed to see that any window of opportunity was eliminated by what they themselves were without demur recording as the sacristan’s actual proposition: that during the whole “five or six minutes” the sacristy door remained locked.
If the Judgment could go wrong in these ways, and not notice its own obvious errors, how much more so could the second jury.
John Finnis AC QC is professor emeritus at Oxford University, having been Professor of Law and Legal Philosophy from 1989 to 2010. He is a Fellow of the British Academy (Law and Philosophy sections). A barrister of Gray’s Inn, he practised from 1979 to 1995 and was appointed Queen’s Counsel [QC] (honoris causa) in 2017. Originally from South Australia, he was created a Companion in the Order of Australia in 2019 ‘for eminent service to the law, and to education, to legal theory and philosophical enquiry, and as a leading jurist, academic and author’.