She Invented the Threats. He Was Arrested Twice.

On March 17, 2020, police in the Australian state of Victoria arrested Bryce Cadby for a second time. His former partner had supplied messages that appeared to show him threatening her. An intervention order was already in place. Cadby denied sending the messages and said he had complied with its conditions.

Officers examined his phone. They became concerned that the messages might have been fabricated.

Eight days later, another complaint from the same woman generated an application to remand him.

Those two events appear in the sentencing reasons in DPP v Bacon, a case decided in the County Court of Victoria in September 2022. Their proximity supplies the central question in this story: what happened to the doubt recorded on March 17 when the next allegation arrived? The judgment establishes the sequence. It does not explain the internal police decisions connecting it. Sentencing reasons, paragraphs 17 to 22.

The woman was Makaela Bacon. She eventually pleaded guilty to five counts of perjury and a related charge of making a false report to police. Judge Peter Rozen convicted her and imposed an 18-month community correction order. The sentence required supervision and assessment for treatment and programs. It included judicial monitoring. She received no prison term. Sentencing reasons, paragraphs 1 to 3 and 90 to 97.

Cadby had been arrested twice and subjected to a family violence intervention order because of an accusation campaign that the court described as serious, planned offending. According to the judgment, he was never charged with a criminal offence or placed on trial. That distinction matters. His case demonstrates how much state intervention can occur before anyone has to prove guilt in a criminal courtroom.

It also demonstrates how an investigation can recover. Police checked his phone, examined hers, obtained telephone records and stopped investigating him after her admission and the corroborating analysis. Bacon’s prosecution followed. A fair account must include those steps alongside the arrests.

Unchecked reviewed the published sentencing reasons for this article. The document contains an agreed account of the offending and the judge’s explanation of the sentence. It is substantial, but it is not the complete police file. It does not contain the original interviews, forensic extractions, intervention order file or internal communications. Where those missing materials would be necessary to establish a failure by an officer or agency, the distinction is made explicit.

What the available record already establishes is disturbing enough. A fabricated account of violence acquired the authority of a court order. Fabricated messages then appeared to show defiance of that order. Each new complaint could make the previous one seem more credible, while making the accused appear more dangerous.

The lies ultimately failed. Before they failed, public authority gave them consequences.

1. The allegation that started the machinery

Bacon and Cadby had known each other for approximately two years. Their relationship began in late December 2019. On February 24, 2020, Bacon attended Werribee Police Station and reported an incident of family violence that she said had occurred the previous evening.

Her account began with an ordinary visit: time together at his home, television, physical affection. It then turned into a claim that he had reacted violently when she refused greater intimacy and suggested taking a break. She said he choked her until she lost consciousness, after which she recovered and left.

She signed a statement acknowledging both its truth and the penalties for making a false statement. In the later criminal proceedings, that statement formed the first perjury count. The sentencing judge recorded that she signed it knowing it was false. Sentencing reasons, paragraphs 4 to 6.

The allegation required a serious response. A claim of violence causing loss of consciousness presents an obvious safety concern. Police cannot establish an allegation’s truth by treating its seriousness as proof, but they also cannot postpone every protective decision until a criminal trial. The difficult work lies between those positions: establishing what can be checked immediately, protecting people from credible danger, and keeping uncertainty visible as decisions are made.

On February 25, officers arrested and interviewed Cadby at Pakenham Police Station. He denied the allegations. They observed that he walked with a substantial limp. He told them he had an acquired brain injury that effectively prevented him from moving half his body. He was served with a complaint and warrant for a family violence intervention order and bailed to appear in the magistrates’ court two days later. Sentencing reasons, paragraph 7.

His disability was information to investigate. It did not, by itself, establish that he could not commit violence. It did require attention to the specific movements and physical actions being alleged. An investigator assessing an account of an assault should want to know whether the described sequence fits the people involved and the environment in which it supposedly occurred.

The available judgment does not establish how officers evaluated that issue at the time. It records their observation and his explanation. It does not supply a medical assessment of his capabilities or the arresting officers’ reasoning. Calling the arrest obviously unlawful on this record would go beyond the evidence.

The complaint nevertheless had immediate effects. Cadby had to respond to police and then to legal proceedings. He occupied the position of a person accused of family violence before the evidence behind the accusation had been resolved. The intervention process added obligations and potential consequences of its own.

That is the first institutional feature worth examining. An accusation can generate several separate decisions, each made by a different person with a different task. One officer takes a statement. Others arrest and interview. A court considers an interim order. Later officers may encounter the order as an established fact in the background of another report.

The danger is that the existence of those decisions can gradually substitute for examination of the original evidence. An order is real. An arrest is real. Neither makes a fabricated underlying allegation true.

In this case, that distinction would become especially important. The order did more than regulate contact. It provided the setting for the next stage of Bacon’s offending, in which she supplied supposed evidence that Cadby was ignoring the court and continuing to threaten her.

2. Marks, medical evidence and a changing account

Bacon underwent a medical assessment on February 25 as part of the police investigation. Photographs were taken and her neck was examined by CT scan. The judgment records a normal scan with no apparent acute injury. It also records that marks extending around her neck were considered inconsistent with bruising caused by strangulation.

The following day, she gave police a further statement. She added an allegation that Cadby had followed her as she left his bedroom, pushed her from behind on the stairs and caused her to fall. She said she hit her head but was not seriously injured. That statement became the second perjury count. Sentencing reasons, paragraphs 8 to 10.

The medical findings belong in the account because the court included them in its factual reconstruction. Their significance should remain specific to that evidence. A reader should not take a normal scan, or a particular injury pattern in one case, as a general test for whether an assault occurred. Here, the finding of fabrication ultimately rests on the admitted offences and the agreed facts, supported by the wider investigation.

There is a similar distinction concerning the second statement. Additional detail is not inherently dishonest. A witness may remember more, explain an event more clearly, or answer questions not asked during an initial interview. The relevant issue is whether a material addition withstands examination. Bacon’s account of being pushed down the stairs did not. She later admitted that Cadby had not pushed her.

What makes this sequence consequential is the opportunity for investigation created by the new information. By February 26, the file contained a serious allegation, a denial, an observed physical impairment, medical findings and an additional account of violence. Those were distinct pieces of information. They needed to be compared, rather than assembled automatically into a single narrative of escalating abuse.

The sentencing document does not tell us when every officer received the medical findings. It does not say whether the person taking the second statement had the photographs, scan result or an expert explanation in front of them. It does not identify the evidence placed before the magistrate the next day. These omissions prevent a conclusion that particular officers knowingly ignored exculpatory material.

They also identify useful questions for an institutional review. When was the medical assessment available? Who reviewed it? Was the distinction between observed injury and the alleged cause of injury preserved in the police summary? Did the intervention order application accurately present the unresolved evidence? Was anyone responsible for bringing the material together?

Those questions are more exacting than asking whether police should have believed Bacon. Belief describes an attitude. An evidence record can show a decision, its basis and its timing.

On February 27, an interim family violence intervention order was granted in the Dandenong Magistrates’ Court. Bacon was listed as the affected family member and Cadby as the respondent. Sentencing reasons, paragraph 11.

For the next stage of the case, the legal setting had changed. Bacon could report an alleged breach of an existing order. Such a report carried a different implication from the initial accusation: a person already subject to court intervention was supposedly continuing his conduct.

That implication depended on the authenticity of the new evidence. Without authentication, the order could become a source of borrowed credibility for another lie.

3. A threat with a name attached

On March 7, Bacon returned to Werribee Police Station with a screenshot. It appeared to show a Facebook message from someone identified as Bryce. The message threatened to kill her and referred back to the earlier supposed incident.

She did not provide a formal statement on this occasion. Even so, the report was recorded as an intervention order breach, and police made arrangements to arrest Cadby. The later sentencing reasons identify this as the separate offence of making a false report to police. Sentencing reasons, paragraphs 12 and 13.

Two days later, Bacon returned with screenshots of more than ten messages. Their content suggested possession, threats to her family, and plans for severe violence. She said she had not replied because she feared Cadby would come to her home and carry out the threats. She signed another statement knowing it was false. That was the third perjury count. Sentencing reasons, paragraphs 14 to 16.

A screenshot has an immediate persuasive force. It seems to preserve a moment rather than merely describe one. The reader sees words, an apparent sender and the familiar presentation of a conversation. The image appears to answer the question of whether something was said.

It cannot answer every question about its own origin. The text visible on an image and the identity of the person who authored it are separate matters. In a criminal investigation, the connection between them is central. Here, the content was frightening enough to increase urgency precisely because it purported to be Cadby’s own speech.

The messages also did more than threaten future harm. They referred to supposed earlier violence. That gave them a second evidentiary function: they could be read as admissions. A fabricated message can make an original accusation appear corroborated by the accused, even though both the accusation and its apparent confirmation come from the same source.

The court’s account makes that mechanism visible without requiring a theory about how often it occurs. Bacon created messages and attributed them to Cadby. She then presented them as evidence of his conduct. The apparent corroboration was circular.

The criminal significance of the order added another layer. If authentic, the messages could indicate continued danger and disobedience of a judicial instruction. The same image could therefore be invoked to support several propositions: that he had abused her, that he remained threatening, and that prior intervention had failed to restrain him.

This is why authentication is an essential part of protection. A forged threat can point official attention at the wrong person and obscure the actual conduct occurring between the parties. Checking a message’s origin does not diminish the seriousness of a genuine threat. It establishes whose threat it is.

Nothing in the sentencing reasons establishes the exact checks performed on the March 7 and March 9 material before the second arrest. The record is clear about what happened when police subsequently examined Cadby’s phone. On March 17, during his second arrest and interview, he denied sending the messages and said he had complied with the order. The phone review caused officers to suspect fabrication.

That was a significant correction in the investigation. His account was tested against a device rather than assessed solely through the accusations already in the file. Information capable of weakening the case was identified by police themselves.

But a suspicion recorded during one interview can protect someone only if it affects later decisions. The next part of the chronology raises the question of how far this one travelled.

4. Eight days between doubt and a remand application

Following the March 17 interview, Cadby was released pending summons. On March 25, Bacon made another report at Werribee. This time, the screenshot she produced purported to combine threats with an admission of rape. She gave another signed statement. It became the fourth perjury count.

The judgment then records that an application to remand Cadby was sent to Pakenham Police Station. Sentencing reasons, paragraphs 17 to 21.

The document does not say that a magistrate remanded him. It does not establish a third arrest or a period of imprisonment. It records the transmission of an application. Preserving that limitation is essential to understanding what the source actually proves.

Even within that limit, the sequence deserves attention. One part of the police response had generated doubts about the authenticity of messages attributed to Cadby. A subsequent complaint involving another message was followed by a step directed towards custody. The reasons for moving from one position to the other are absent from the published account.

There are several possible explanations, and the judgment does not choose among them. The later decision makers may not have received the earlier concern. They may have received it and considered the new material distinguishable. There may have been other information in the file. An application may have been subject to review before any further action. Each possibility would lead to a different assessment of institutional performance.

An accountability investigation should obtain the documents that distinguish those explanations. The relevant material includes the March 17 interview record, the notation arising from the phone examination, the March 25 statement, the remand application and any supervisor’s assessment. The question is whether the earlier doubt was communicated, weighed and resolved before further coercive action was considered.

This approach does not require assigning a dishonest motive to an officer. A system can expose someone to avoidable harm through incomplete communication, unclear responsibility or a poorly structured file. Those failures would matter even if every person involved believed they were carrying out a legitimate task.

The use of different stations makes the information question particularly concrete. Bacon made reports at Werribee. Cadby was interviewed at Pakenham. The interim order was granted in Dandenong. The published reasons identify those locations, but do not describe how investigative responsibility was allocated between them.

For a complainant experiencing genuine danger, coordination can determine whether protective information is acted upon promptly. For someone facing fabricated allegations, coordination can determine whether exculpatory information prevents another unnecessary intervention. In both situations, important facts must reach the people making the next decision.

The case therefore supplies a narrowly defined public interest question: what process ensured that the concern about fabricated messages was visible across the investigation?

The eventual perjury convictions do not answer it. Convicting the person who supplied false evidence resolves her criminal responsibility. It does not automatically audit the institutions through which the evidence passed.

Nor does the published chronology establish that police failed that test. It establishes the need to inspect the underlying record before claiming either institutional failure or an adequate response. The difference between those conclusions is precisely what a serious follow-up should investigate.

The following day brought a more direct examination of Bacon’s evidence. She consented to an electronic analysis of her phone. That examination would expose a feature of the fabrication that was more concrete than any argument about her credibility.

5. The evidence inside the phone

The initial examination on March 26 identified image files on Bacon’s phone with the title “Fake Texts.” Police returned the phone while awaiting the results of call record analysis.

On March 27, knowing that police were examining her telephone records, Bacon returned with another allegation. The material now included apparent pictures of her house and a knife, accompanied by a threatening message. She signed a fifth false statement. Sentencing reasons, paragraphs 22 to 25.

The escalating content is significant. The earlier material had threatened harm. The newer material appeared to bring the supposed threat closer to action by associating it with a home and a weapon. For someone assessing immediate danger, that presentation could demand a more urgent response.

For someone assessing authenticity, the question remained the same: where had this material come from, and what connected it to Cadby?

The discovery on the phone created another important point in the chronology. A filename alone does not establish an entire offence. It is, however, information that an investigator can examine alongside the earlier doubts and the original files. By this stage, police had a concrete basis for testing whether the images were manufactured rather than records of messages actually received.

The judgment does not explain why the phone was returned when it was, what had been copied, or what further analysis remained possible. There is no basis here to assert that officers lost or destroyed evidence. Those details would require the extraction report and the associated handling records.

On April 1, Bacon called police and admitted that she had fabricated the messages. Telephone record analysis confirmed no contact between the pair apart from one unanswered call from Cadby to her on March 17. Police then ceased all investigations concerning him. Sentencing reasons, paragraph 26.

The exception for that unanswered call should remain in any accurate account. Saying that there had been absolutely no contact would misstate the judge’s description. The significance of the records was that they did not support the manufactured exchanges alleged in the complaints.

The case was now supported by more than a retraction. There was an admission of fabrication, the earlier examination of Cadby’s phone, material found on Bacon’s device and the record analysis. These sources supplied a way to evaluate the claim of fabrication independently of the word of either party.

That combination also explains why the case cannot fairly be reduced to a general argument that women who withdraw allegations must have lied. Withdrawal and fabrication are different propositions. Bacon’s case proceeded through an investigation and guilty pleas to identified offences. The basis for describing these allegations as false is specific and documented.

The investigation’s success deserves recognition. Police did identify a problem with the evidence and pursued it. They stopped investigating Cadby once the admission and analysis established what had occurred. The criminal justice system later convicted the person who supplied the false statements.

At the same time, successful detection at the end leaves earlier decisions open to scrutiny. The interval between the first suspicion and the closure of the investigation mattered to the person still exposed to it. An agency can deserve credit for correcting course while being asked whether it could have done so more quickly or with fewer unnecessary consequences.

That is the useful standard for reviewing this case. The question is how an evidence check became a change in institutional action, and whether the route between the two was clear enough to protect Cadby while the remaining work was completed.

6. An admission with a justification attached

Police arrested Bacon on April 8 and interviewed her. Her answers did not amount to a simple, uninterrupted confession to every aspect of the offending. The judgment records admissions alongside claims she continued to make.

She admitted manufacturing the messages and identified an application she had used. She acknowledged that Cadby had not breached the intervention order and had not pushed her down the stairs. She also said she had fabricated the messages because he had committed an offence against her and she thought alleged breaches would cause the matter to be taken more seriously. At that stage, she maintained that the substance of the messages was true and that some marks on her neck came from his hands. Sentencing reasons, paragraph 27.

Those were her statements during the interview. They should not be reported as the court’s acceptance that the underlying accusations were true. The eventual sentencing proceeded on guilty pleas and an agreed prosecution account. The judge expressly recorded the knowing falsity of the statements constituting the perjury counts.

It is equally important to preserve the interview account because it reveals the distinction Bacon herself attempted to draw. She conceded the creation of evidence while seeking to justify it by reference to wrongdoing she still alleged. The record of that attempted justification is part of the history of the investigation.

Fabricated corroboration is dangerous because it tries to transfer a disputed assertion into another person’s mouth. A police officer considering an accusation knows it is an accusation. A message apparently written by the accused can look like independent support, particularly if it includes an apparent admission.

In this case, the manufactured communications purported to provide precisely that support. They presented the alleged perpetrator as acknowledging earlier conduct while threatening more. Once the authorship was exposed, the messages could no longer perform either function.

The broader evidentiary principle is straightforward. A person’s conviction that they have been wronged does not authenticate evidence they create to support that conviction. Investigators must still establish what happened, and a complainant cannot supply a fictitious admission from the accused to resolve the issue.

The repeated signed statements added another dimension. These were documents presented to police with express acknowledgements about truth and the penalties of perjury. Bacon admitted that an officer had explained those penalties to her. The court later treated the offending as a sequence requiring planning rather than a single spontaneous lie in a difficult moment. Sentencing reasons, paragraphs 27 and 51 to 54.

The progression matters when assessing responsibility. There were repeated opportunities to stop. The material changed. The apparent threats grew more elaborate. New reports drew on the legal position created by earlier reports. A system designed to respond to repeated danger was being supplied with invented repetition.

Understanding that conduct does not require a diagnosis from a commentator. The court had psychiatric evidence and made findings about how Bacon’s mental functioning affected her culpability. Those findings belong in the explanation of her sentence. They do not make the manufactured messages authentic or erase the decisions they prompted.

A complete account therefore has to keep several questions distinct. What did she do? What did she know about the falsity of the statements? What role did her mental condition play? What punishment was justified? What should institutions learn from the way the evidence moved through the system?

The first questions were addressed in the criminal proceedings. The final one requires an examination of institutional conduct that the sentencing exercise was not designed to provide.

7. What happened to Bryce Cadby

The published reasons identify Cadby as the direct victim of Bacon’s offending. Judge Rozen recorded that he had been subjected to an unwarranted investigation, arrested twice, made subject to an intervention order and then accused of violating it through violent messages.

Cadby did not submit a victim impact statement. The judge nevertheless recognised that the experience must have caused stress and anxiety. The reasons also expressly state that Cadby was not criminally charged or put on trial. Sentencing reasons, paragraphs 61 and 62.

This leaves a real limit on what can be said about his personal experience. The document does not establish lost wages, legal fees, a lost job, the collapse of another relationship or a diagnosed psychiatric injury. No such details should be supplied to make the story more dramatic. Nor does the absence of a victim impact statement tell us why he did not provide one.

The documented interventions are substantial in their own right. Arrest requires a person to submit to state authority. An interview demands an answer to allegations with potentially serious consequences. An intervention order changes the legal circumstances in which contact and behaviour will be assessed. A later complaint can create fresh exposure even while the original accusation remains disputed.

Cadby’s case makes the period before charge visible. Public discussion often measures a miscarriage of justice by the length of a prison term or the reversal of a conviction. Those are vital measures. They cannot capture every harm created by a false allegation. A person can be correctly spared prosecution and still have undergone coercive action because someone deliberately supplied false information.

His disability adds a further question about how the response was adapted to him. The judgment records an acquired brain injury and an observed limp. It does not establish what assistance he required during arrest or interview, what was offered, or whether his account of his physical limitations was independently evaluated.

An appropriate follow-up would ask those questions without presuming their answers. It would also avoid treating disability as a reason to dismiss the possibility of violence in every case. The relevant issue is the individual’s actual capacities and needs, examined against the specific allegation.

There is a separate question about the legal record left behind. The sentencing reasons say that police investigations concerning Cadby stopped on April 1. They do not provide the date or terms of the intervention order’s discharge, if and when it was discharged. They do not describe the correction of police databases or information supplied to other agencies.

Closure of an investigation and correction of every record arising from it are distinct administrative tasks. A clear account of the outcome should identify which records were updated and how the false allegations were marked. That matters because the original documents may otherwise remain available to someone making a later decision without the final finding immediately beside them.

This is an accountability question, not an assertion that Cadby’s records were mishandled. The materials reviewed do not establish that they were. They establish why the answer should be sought.

Any attempt to speak with Cadby should begin by asking whether he wants to participate. Public interest in a case does not create an obligation for the person harmed to narrate it again. The existing judgment is sufficient to report the established offending. An interview could add his account of its effects, provided he chose to give it and the reporting distinguished recollection from independently verified facts.

The lack of that interview should remain visible. A publication claiming to stand up for people harmed by institutions should be especially careful not to invent a victim’s voice.

8. Why the court did not send Bacon to prison

The absence of a prison sentence is likely to be the first thing many readers notice. Five perjury counts, repeated false reports, two arrests of an innocent person, and an 18-month community correction order: the result invites scrutiny.

The sentencing reasons provide an explanation considerably more detailed than a headline about escaping jail.

Bacon was 19 when she offended and 21 at sentencing. She had no prior criminal history. The court considered two reports from forensic psychiatrist Dr Nina Zimmerman, treatment records, a report from her community mental health service, references, a Corrections Victoria assessment and a court mental health assessment. Sentencing reasons, paragraphs 28 to 50 and 68 to 71.

The psychiatric history predated the offending. It included significant mental health difficulties and inpatient treatment. Zimmerman diagnosed borderline personality disorder and identified a direct relationship between Bacon’s condition and the conduct at issue. She described impaired emotional regulation and an altered experience of interpersonal events under stress.

These were conclusions about Bacon based on the material before the court. They are not a basis for treating other people with the same diagnosis as dishonest or dangerous. The relevance was individual: how her mental functioning affected this offending and how a sentence would affect her.

Judge Rozen accepted that the evidence engaged the sentencing principles associated with Verdins. As he explained them, impaired mental functioning can reduce moral culpability while leaving legal responsibility intact. It can also affect the appropriate kind of sentence, the weight given to deterrence and the consequences of imprisonment for the individual. Sentencing reasons, paragraphs 55 to 60 and 72 to 77.

That distinction is important to the result. Bacon was convicted. The court did not decide that she bore no responsibility. It decided that the punishment had to reflect the established relationship between her mental condition and her offending, together with the other sentencing factors.

The judge accepted evidence that imprisonment would affect her more severely than a person without the relevant condition and risk a significant deterioration in her mental health. He also considered the value of maintaining her treatment. Her youth gave rehabilitation particular weight.

The offending was already about two and a half years old by the time of sentence. The judge said the delay was not her fault, that there was no evidence of further offending in the intervening period, and that the reports indicated compliance with treatment. Those were findings as of September 2022. They do not establish her subsequent history. Sentencing reasons, paragraphs 78 to 85.

Her guilty pleas also mattered. The judge made the formal declaration that, without them, he would have imposed six months’ imprisonment followed by an 18-month community correction order. That statement explains the effect of the pleas in this case. It should not be presented as a universal tariff for a false accusation or as a punishment she actually served. Sentencing reasons, paragraph 97.

The prosecution accepted that a community correction order was an available disposition in the circumstances. That is narrower than saying prosecutors believed the conduct was trivial, and narrower than saying they demanded the exact order ultimately imposed. The judge recorded the defence’s submission for such an order and the prosecution’s acceptance of its availability. Sentencing reasons, paragraphs 65 to 67.

The decision remains open to reasoned criticism. A reader may give greater weight to the repetition, planning and impact on Cadby. But criticism should engage the explanation actually given. Claiming the judge simply did not care about false allegations would ignore his express assessment that the offending was grave and that punishment had a role.

The harder question is whether the sentence’s promised combination of treatment, supervision and protection was delivered. A reasoned decision is the beginning of that inquiry.

9. The conditions behind the sentence

There was a difference between two risk assessments recorded in the judgment. Zimmerman assessed Bacon as presenting a low risk of acting harmfully towards others. Corrections Victoria assessed a medium risk of reoffending. The assessments addressed differently expressed questions and may have used different methods. The reasons do not supply enough information to declare one assessment wrong or to treat them as directly interchangeable. Sentencing reasons, paragraphs 47, 69 and 82.

The difference is still relevant to understanding the order. A finding of good rehabilitation prospects did not mean the court considered all future risk absent. Supervision was part of the response. The court’s mental health service also assessed Bacon, endorsed a treatment condition and identified no barriers to effective management by community corrections.

The judge expressed a specific concern about her ambivalence towards working with a psychologist. The psychiatric evidence recommended specialised therapy, while recognising questions about her willingness and readiness to engage. Her case worker indicated that she might now be more willing. The judge proposed that the psychiatric reports and sentencing reasons be supplied to Corrections Victoria and that treatment progress be monitored through reports to the court. Sentencing reasons, paragraphs 68 to 71 and 90 to 91.

This is a useful point of scrutiny because it is tied to something the judge actually identified. The success of the sentencing approach depended partly on the availability of treatment, engagement with it and a supervisory structure that could respond if things went wrong.

The final order required supervision for its duration. Bacon was to undergo assessment for mental health treatment and, if suitable, complete recommended treatment and counselling. She was also to be assessed for and complete programs recommended by Corrections to reduce offending risk. Judge Rozen directed her to return for judicial monitoring on December 12, 2022.

Standard conditions included reporting to community corrections, notifying relevant changes of address, complying with visits and instructions, and obtaining permission before leaving Victoria. The judge warned that breach could return her to court and lead to resentencing for the original offending as well as a sentence for the breach. Sentencing reasons, paragraphs 92 to 96.

A description that says nothing happened to Bacon would therefore be inaccurate. She received convictions and an enforceable order. Whether that was a sufficient response is a separate judgment, which can be debated without erasing the conditions imposed.

There is also a difference between imposing conditions and demonstrating their effect. The sentencing reasons record the intended arrangement. They do not report the December hearing or establish whether she subsequently completed the order, breached it, received every recommended service or returned to court. Those outcomes have not been verified for this article.

General research explains why implementation deserves attention, without resolving Bacon’s individual case. A 2017 Sentencing Advisory Council study followed 7,645 people given community correction orders in Victoria during a year beginning in July 2012. Around half breached an order through further offending or another form of noncompliance. The council stressed the importance of early support. That older, mixed cohort cannot supply a prediction about Bacon. Sentencing Advisory Council, research announcement.

A separate study by Rachael Green, David Hopkins and Garry Roach used surveys of 200 adults on such orders and interviews with 20 participants in outer western Melbourne. Its published abstract reported limited evidence that support addressed several needs relevant to rehabilitation. That finding supports asking what services were delivered in an individual case. It does not establish a service failure here. Study abstract, 2020.

The follow-up should therefore seek the publicly available outcome of judicial monitoring and any later orders, while respecting the privacy of clinical treatment records. Readers can reasonably ask whether the court’s directions were carried out. They do not need access to every confidential therapy session to assess whether the justice system completed the work it promised.

10. Protection depends on getting the evidence right

Judge Rozen identified a wider public cost in Bacon’s offending. Police resources had been diverted to false complaints. He also discussed the possibility that such cases might make police slower to respond to genuine family violence allegations. Crucially, he said there was no evidence that this had happened as a result of Bacon’s conduct and did not count that possible effect when sentencing her. Sentencing reasons, paragraphs 63 and 64.

That restraint provides a useful model for reporting the case. It is possible to identify a risk without declaring an unproven consequence. The record establishes diversion of resources and harm to Cadby. It does not establish a later victim whose complaint was mishandled because an officer had encountered Bacon’s lies.

The same care should govern claims about the frequency of false accusations. One documented prosecution cannot provide a population rate. It cannot establish that a particular unrelated complainant is lying or that a whole category of complaints is unreliable. Its value lies in showing a specific mechanism and identifying decisions that can be reviewed.

Victoria Police’s later published code recognises that misidentifying the predominant aggressor can cause serious harm. It discusses corrective action and directs concerns to a family violence liaison officer or station supervisor. The code also describes collecting physical and medical evidence and supervisory assessment of whether a prosecution has sufficient support. These provisions are in the 2022 edition, which postdates Bacon’s offending. They should not be treated as proof of the precise requirements governing every decision made in early 2020. Victoria Police code, sections 2.1.6.2, 4.1.2 and 4.2.7.

The present public description of family violence liaison officers assigns them responsibilities for quality assurance of reports and evidence briefs, coordination and assistance to frontline police. It also identifies support for people with disabilities among their responsibilities. That describes an institutional point at which questions about current safeguards can be directed. Victoria Police, liaison officer responsibilities.

An effective safeguard would connect a doubt about evidence to the next decision that relies on it. If a phone examination raises concern that a message was fabricated, the concern should be accessible to the person assessing a later complaint about similar messages. The record should make clear whether authenticity remains unresolved and what work is underway to resolve it.

This is an editorial recommendation drawn from the chronology, rather than a finding that a particular policy was breached. It would be tested by inspecting the actual file. A useful review would ask whether the system preserved the distinction between a complainant’s description, an officer’s observation, an expert conclusion and a court’s interim action.

The distinction matters at each stage. An officer can observe a mark without knowing its cause. A court can make a protective order before the disputed facts have been fully tested. A screenshot can display threatening words without establishing who wrote them. Each item can be important while remaining limited.

Good investigation keeps those limits attached to the information. Otherwise, a sequence of individually qualified facts can become an apparently certain account as it passes from one document to another. The person reading the final summary may not see the uncertainty that accompanied the first observation.

The response also needs a way to correct direction promptly. When an allegation is disproved, the agency should identify decisions still resting on it and assess what needs to change. The correction should reach the operational record as reliably as the original complaint did.

The available sources do not establish how Victoria Police performed each of those tasks in Cadby’s case. They do establish an investigation in which fabricated evidence generated repeated official action before being exposed. That is sufficient reason to ask for a precise explanation of the controls, their use and any lessons drawn afterwards.

The public does not benefit from a choice between indifference to domestic violence and indifference to evidence. The people seeking protection, and the people facing accusation, both depend on an institution capable of distinguishing what is alleged from what is established.

11. What a complete reckoning would require

There are three sets of records that would move this story forward.

The first concerns the investigation. The original statements and medical assessment would show how the initial case was presented. The phone examination records would establish the timing and scope of the doubts. The remand application would show what was being requested after those doubts arose, on whose authority and on what evidentiary basis.

The second concerns the intervention order and the correction of the record. The court file would establish the order’s conditions, duration and eventual disposition. Police records could show how the false allegations and the cessation of the investigation were documented, including any communication to agencies that had received the original account.

The third concerns the sentence. Publicly accessible monitoring records or subsequent decisions could establish what happened after September 2022. They could show whether the supervised rehabilitation contemplated by the judge proceeded as directed, without disclosing confidential clinical information unrelated to the public question.

Obtaining those materials might strengthen criticism of an institution. It might answer a concern in that institution’s favour. Either result would be useful. Accountability reporting should be capable of recording a competent response as clearly as a failure.

The questions to police are particularly focused. Who was responsible for the investigation as a whole? How was the concern arising from Cadby’s phone recorded? Was it available to the people handling Bacon’s later reports? What assessment accompanied the remand application? What additional safeguards, if any, followed the case?

The questions to corrections concern implementation. Was the judicial monitoring process carried out? What can be publicly confirmed about completion or enforcement of the order? If particular information cannot be released, what general evidence is available about the operation of the supervision and treatment arrangements on which the sentence relied?

The questions for the parties should be offered as an opportunity to add verifiable information. Cadby may be able to explain the personal effects and identify records the judgment does not contain. Bacon or her representative may be able to provide subsequent court documents or address how the public account should be updated. Their participation would not alter the existing convictions without a legal basis, but it could make the account more complete.

Unchecked has not sent those questions for this article. No refusal to comment is implied. There is no basis to say an agency has stonewalled a request it has not received.

The established case already warrants publication as an analysis of the court record. It shows a woman repeatedly supplying false allegations and manufactured evidence, a man subjected to two arrests and an intervention order, and investigators eventually identifying the fabrication. It also shows a court imposing a sentence designed around both the seriousness of the conduct and the offender’s individual mental health circumstances.

The unresolved institutional questions should remain attached to that account. A perjury conviction identifies the person responsible for the false statements. It does not explain every decision taken on the strength of those statements or demonstrate that every resulting record was corrected.

For Cadby, the essential event was the end of an investigation that should never have been directed at him on the basis of fabricated conduct. For the institutions involved, the essential task extends further: establish how the evidence was checked, how doubts were shared and how the consequences were repaired.

The file contains a moment when police began to see that the messages might be false. It contains a later step towards remand. The documents between those moments would tell the public something the sentencing judgment cannot: whether a warning identified inside the system was able to restrain the system’s power.

Explore the Record: The Record: Makaela Bacon / Bryce Cadby