Chapter 19 — Building the Evidence

When memory could not always remain continuously accessible—and institutional records could not be trusted merely because they were official—I needed an external structure capable of preserving sources, dates, changes, contradictions and the relationships between them.

Building the evidence did not begin as a technical project. It began with the need to survive the repeated demand to tell the story again.

Every new lawyer, clinician, agency or complaint process wanted its own account. Each retelling carried a cost. Dissociative amnesia meant that material available to me on one day might be difficult or impossible to reach on another. If the person receiving the account misunderstood it, left their position or failed to preserve it accurately, the work could be lost.

I needed something outside my own mind which could retain what had been available at a particular time.

That need eventually became the Evidence Map.

Preserve First, Interpret Second

The first principle was simple: preserve the source before trying to explain it.

An email should remain an email, with its sender, recipient, date, subject and attachments. A medical report should be retained as the document supplied, not merely as a quotation copied into my notes. An audio recording should remain available in its original form even after a transcript had been made. A website page should have a dated copy so that later revision did not erase what had been published earlier.

Interpretation could then be placed beside the source rather than silently replacing it.

This distinction protected the evidence from my own later conclusions as well as from institutional rewriting. If I changed my understanding, the original document remained available to show what had actually existed before the change.

Different Kinds of Evidence

Not every item performed the same task. I began separating material into different evidentiary categories.

There was direct recollection: what I remembered experiencing. There was a contemporaneous account: what I wrote or said at a particular time. There was information reported by another person. There was an institutional record. There was physical or clinical evidence. There was independent corroboration. Finally, there was inference: a conclusion I reached by comparing two or more of those sources.

The categories could overlap, but they should not be confused. A statement in a hospital file might prove that a clinician wrote something on a particular date without proving that the statement itself was correct. An email might establish that I disclosed a name years earlier without independently proving every event associated with that name.

An inference could be strong, especially where several independent sources converged. It still needed to remain identified as an inference rather than quietly becoming a memory.

Dates as Evidence

Dates helped answer one of the recurring attacks upon my credibility: the claim that a disclosure had appeared only after a later crisis, conversation or publication.

The Sandy Collins report dated 4 April 2005 established that significant disclosure and trauma-related difficulty were already being discussed by then. Later correspondence showed the development of names, questions and allegations across time. Website publication preserved another layer of what I was able to state at particular stages.

A date did not prove the historical event described. It established that the account or concern existed by that date. That could be decisive when somebody later claimed that the material had only recently been invented.

File dates also required caution. The date a copy was downloaded or scanned was not necessarily the date the document was created. An email's sent date, a report's stated date, a filesystem timestamp and the date on which I received the item could all be different. The evidence structure needed to preserve those distinctions.

Names Before Later Denial

The 2014 Ballarat CASA assessment prepared by Andrea Lockhart became an important stable marker. It recorded the names Sister Ursula, Sister Vincent, Sister Germaine and Sister Genevieve before later assertions that I had been unable to identify the women involved.

The document does not prove every remembered event associated with those names. It proves that the names were recorded in an independently authored professional document at that time. Any later claim that I had never named them must be examined against that earlier record.

This is the kind of narrow proposition evidence can establish strongly. It does not need to carry the whole history in order to invalidate a later categorical assertion.

Correspondence as a Timeline

Emails and letters allowed a timeline to be reconstructed from ordinary exchanges. They showed whom I contacted, what I asked, what material I supplied, what answer was given and where no answer came.

An unanswered email was not proof of why somebody failed to respond. It was evidence that a dated request had been made and that no response had been located. A sequence of unanswered requests following a disputed professional action could show procedural abandonment even where the institution never formally announced that it had abandoned the matter.

Correspondence also preserved changing explanations. Where a person first described a close relationship and later minimised it, the earlier words remained available. Where an organisation gave one reason for a decision and later relied upon another, both versions could be placed side by side.

The value lay in the sequence. A single message could be ambiguous. Twenty messages arranged by date could reveal a pattern of knowledge, delay, redirection or avoidance.

Recordings and Transcripts

Recordings became necessary because a meeting could be important while my access to it afterwards was incomplete.

With dissociative amnesia, being present for a conversation did not guarantee that I could later reconstruct it reliably. Stress, medication and the subject being discussed could further affect retention. A recording allowed me to return to the actual words rather than depend upon whichever fragments remained accessible.

Where recordings were lawfully made or obtained, I preserved the original audio. A transcript made the content searchable and easier to compare, but it remained a derivative record. Names could be misheard, overlapping voices could be attributed incorrectly and pauses or tone could be lost. Important passages therefore had to be checked against the sound itself.

The recording also protected other participants. It reduced the risk that I would unknowingly attribute words to somebody which they had not used. Its purpose was accuracy, not simply support for my preferred account.

The Website as External Memory

My websites became more than places of publication. They were external memory structures.

A page could retain a name, document, argument or connection when I could not rely upon having continuous access to it internally. It allowed material to be assembled progressively rather than forcing me to hold the entire history in mind before writing any of it.

Publication also created dates, audiences and responses. People replied, objected, supplied information or disclosed their own experiences. Those responses became further sources which had to be preserved separately from the page which prompted them.

The capacity to revise a website created its own risk. A correction could remove evidence of what had previously been written. Copies, version histories and dated exports were therefore important. The aim was not to freeze every early error forever, but to make revision transparent.

Following Records Between Institutions

Medical record bundles revealed that documents did not remain within the institution which created them. Material appeared under headings such as Corr In and Corr Out—correspondence received and correspondence sent.

Those labels helped establish transmission. If a disputed psychiatric statement appeared in several systems, the correspondence trail could show whether they had independently reached the same conclusion or merely circulated the same source.

This was central to understanding the false paranoid-schizophrenia label. The number of files containing it did not establish the number of diagnoses. The task was to work backwards through the transfers until the first use could be identified and its actual status examined.

Where the first appearance was a query, an unverified history or an administrative entry rather than a diagnosis, every later copy inherited that weakness even when the question mark or qualification had disappeared.

Reports Within Reports

Professional reports often quoted, summarised or relied upon earlier reports. That could conceal provenance.

A later author might write that I “had” a condition without saying that the information had been copied from a disputed source. A reader could reasonably assume that the later professional had assessed and confirmed it. In reality, no new diagnostic work may have occurred.

The solution was to trace each significant statement backwards. Which earlier record was being cited? Did that record cite another one? Was the original author reporting direct assessment, repeating what somebody else said or merely listing a possibility?

This created a chain of provenance. It showed whether an apparent body of professional agreement rested upon several independent assessments or upon one statement repeated through many documents.

The Disputed Antoce Report

The report associated with Dr Georgiana Antoce required special handling because I dispute its later portions and because the assessment itself extended over approximately five months.

I could not responsibly refer to “the report” as though every page had the same provenance and status. My understanding that approximately the first forty pages reflected Dr Antoce's work had to be kept separate from questions about the authorship, errors and provenance of what followed.

The report also preserved evidence which did not fit the way it was later used. It contained trauma-related symptoms and emerged from sessions in which my attempts at verbal disclosure often became fragmented reenactments. Those observations needed to be considered beside Dr Julian Lim's later diagnosis of dissociative amnesia and the absence of an actual diagnosis of paranoid schizophrenia.

A disputed document should not be discarded merely because parts of it are wrong. It should be disassembled evidentially: who authored each part, what sources were used, what observations remain valid, what conclusions are contested and how later organisations relied upon it.

This principle later became central to my engagement with the lawyer Dr Judy Courtin. In my recollection, she discounted the Antoce report in its entirety rather than separating its evidence, observations, authorship and disputed conclusions. What frightened me was her speaking about going to other psychiatrists until she obtained a report which suited her intended legal purpose.

I understood that approach as being directed towards what would work in court rather than towards establishing the accuracy and validity of what the existing report contained. Dissociative amnesia presented a difficulty for the claim, and it appeared to me that the objective was to remove that difficulty by obtaining a more convenient psychiatric account.

In my opinion, that was a profoundly unprofessional way to approach both the evidence and my claim. The proper response to a complex or disputed report was not to discard everything within it or search for a preferred conclusion. It was to identify what was reliable, investigate what was contested, and obtain further expert opinion through questions directed towards truth and clinical accuracy rather than a predetermined litigation outcome.

My dealings with Dr Courtin require fuller treatment later. This passage records the evidentiary principle against which I believe that engagement must eventually be examined.

Physical and Imaging Evidence

Physical evidence required a similar discipline. Pain, a palpable area or a bodily reaction was real as an experience, but its cause still required investigation.

An ultrasound, CT image, radiology report, clinical examination and history each answered different questions. A report saying that no radiodense foreign body was identified did not necessarily establish that no foreign material existed. It established what the radiologist identified in the area and images available using that method.

The Evidence Map therefore needed to link the written report with the actual imaging, the anatomical location requested, whether that location was fully visualised, earlier clinical observations and any history of fragments being physically removed.

My emotional response to an image or finding also had to remain separate from objective analysis. A reflection could record what the image meant to me; a clinician or appropriately qualified reviewer was still needed to interpret the medical evidence.

The Evidence Map

As the volume increased, folders alone were no longer enough. A document could relate to several people, events, institutions, diagnoses and complaints. Placing it in one folder concealed its other relationships.

The Evidence Map was designed to preserve the source once while connecting it to every relevant part of the history. An event could link to documents, people, locations, physical effects and later institutional responses. A disputed claim could link to its first known appearance, every later repetition, the evidence against it and each correction request.

The map did not decide what was true. It made the basis of a conclusion visible.

That difference mattered. Somebody examining the work should be able to move from a summary statement to the source beneath it, see whether the source was direct or derivative and identify where interpretation had been added.

Chain of Custody

Evidence can be weakened when nobody can explain where a file came from or whether it has changed.

For important documents, I needed to preserve the original filename where possible, the source from which it was obtained, the date received, and any known path through earlier custodians. A working copy could be renamed for clarity, but it should remain linked to the preserved original.

Record bundles required their own history. It mattered whether files came directly from Queensland Health, through a lawyer, from a hospital, from an earlier evidence collection or from a later access request. Two copies of the same report from independent custodians could help establish that the document had circulated, even though they did not constitute two independent reports.

Hashes, timestamps and retained source files could help demonstrate that digital material had not been silently altered. They did not prove that the contents were accurate when first created; they protected the integrity of the copy being examined.

Recording Corrections

An evidence system must allow correction without destroying history.

Names can be misspelled. Dates can be estimated incorrectly. Two incidents can initially be conflated and later separated. Better records can disprove an inference which once appeared reasonable.

The correction should state what changed, why it changed and what source justified the revision. The earlier version should remain recoverable where it has evidentiary importance.

This is particularly necessary for dissociative memory. The ability to refine an account as access improves is part of accuracy. Treating every correction as evidence of dishonesty forces survivors either to preserve known errors or be punished for repairing them.

Evidence of Absence and Absence of Evidence

Missing material also needed to be recorded.

If an institution said that no document could be located, that response became an item of evidence. It established the result of that search at that time. It did not establish that the document had never existed.

Likewise, if a hospital supplied a bundle without a nursing narrative or a lawyer's file lacked an attachment referred to in correspondence, the gap had to be identified rather than silently closed.

The map could then distinguish “not found,” “not supplied,” “known to have existed,” “referred to by another document,” and “no evidence presently located.” Those phrases may sound technical, but each prevents an absence from being made to say more than it can.

Making the Evidence Usable

An archive can contain the truth and still be unusable.

Thousands of pages placed before a lawyer or decision-maker without structure may conceal the decisive document rather than reveal it. I needed timelines, indexes, short summaries and links which allowed somebody to move from the central proposition to the supporting material.

Different readers required different entry points. A clinician might need a concise medical chronology and the exact imaging. A lawyer might need the statement at issue, its provenance, the applicable duty and the consequence. A complaint body might need a short account of the requested correction, the evidence supplied and the response received.

The summary could never become a substitute for the evidence beneath it. Its function was to make the underlying material reachable.

A Record I Could Return To

The Evidence Map had another purpose which was personal rather than forensic. It gave me a record I could return to.

When memory became inaccessible, I did not have to rebuild everything from the beginning. I could see what I had previously recorded, which source supported it and which questions remained open. The system could preserve continuity when I could not depend upon continuity within myself.

That did not transfer authority from me to a database. It allowed the person I had been on one day to leave an accurate structure for the person I would be on another.

Building the evidence was therefore not only preparation for litigation or complaint. It was an accommodation for dissociative amnesia, a protection against institutional rewriting and a method of remaining answerable to the truth as new material emerged.

Once that structure existed, I was no longer limited to arguing against other people's records one document at a time. I could begin reclaiming the record as a whole.