On the record

The trials of court transcription

A YEAR AGO, feeling hopeless about my work as a freelance writer, I began to look for other ways to bring in money – something steadier to tide me over, with possibly even fair pay. One night, I saw a fiftysomething scientist talking on TV about her job dissatisfaction, how she’d left her position, moved to a coastal town with ocean views, and was loving the freedoms of being a remote court transcriber. I thought: maybe I could do that, and while I didn’t move to a beach, I did go onto the website of Australia’s biggest court transcription company to investigate what such work involved.

Within a day, I’d signed up to its training program. Within two days, I began to sit their tests, proving I could tolerate and, more to the point, follow what judges and barristers say to each other, could touch-type at speed, reproduce speech with a 98.2 per cent minimum accuracy rate and obey style manuals for the jurisdictions I would work in. Within two months, I was given ‘live’ work from the courts, which is to say, one morning, while still yawning in my nightie, I opened up my computer in my Melbourne home, put on headphones, located my first audio file – a criminal matter in Brisbane – on the company’s remote transcribers’ site and began to follow the drama unfolding there: a man defending charges of physically assaulting his wife (while his children looked on).

Listening in to what goes on in courts while I’m at home requires the use of specialised software that allows me to download the recorded audio of the court matter to be transcribed. Once that’s done, I can locate the segment of audio I’ve been allocated, which could be at any point in any proceeding – a committal, counsels’ opening or closing addresses in a trial, a judge’s summing up, a Court of Appeal to and fro – and, with the help of a large screen and a foot pedal, I play the audio back, stopping frequently to rewind if I miss or cannot understand or hear a word, and reproduce it as text as fast and accurately as I can. One hour of court audio will take me, roughly, five hours to type.

Mostly, I’m assigned deferred work, which requires me to return my transcript to the company within a day, which will then be distributed to the judge and barristers involved, along with anybody else – such as the jury or the media – granted access to the court record. Daily work is different, requiring a veritable army of transcribers to work together in teams to create transcripts of what’s occurring in court, almost at the same time the matter is unfolding in court.

Such same-day transcripts are complex and often stressful and exacting to work on, and they involve teams of transcribers who can confer with each other as they work, for transcript continuity and coherence, and who are overseen by a supervisor who merges their segments into a seamless whole. Same-day transcripts are returned to the judge and barristers by lunchtime, after the court’s morning session, and by early evening, after the afternoon session. Same-day transcripts are hectic and highly pressured with punishing deadlines; it is transcribing on steroids, and I have only worked on them infrequently, depending on how confident I’m feeling.

But, whichever form of transcribing work you do, the essence of it is the same. Listening in to strangers talk isn’t anything out of the ordinary, of course; most people do this every single day. But listening to frequently verbose legal folk, and those at their mercy, while acting as an authorised, professional eavesdropper is quite another matter. And listening in to these people very, very carefully to register and type each and every last word – rational or otherwise – is, I’ve discovered, an odd and, at times, deeply affecting line of work.


TYPING PRECISELY, TO the letter, what a person says in court – particularly a nervous witness, say, being outfoxed in cross-examination – is a kind of dark art, at times counterintuitive and bewildering, because many people are unaccustomed to telling the truth, let alone the whole truth under oath, and they are prone to excitable, incoherent thinking and its verbal expression. People under stress tend to mangle their syntax. They can struggle to form sentences. Some fail to say very much at all. People who stutter often apologise for stuttering more. Some can utter Freudian slips of the most horrifying or amusing kind, or they head toward very bizarre places, far beyond the constraints of the questions asked of them. My job is to accompany them to these places, to shadow them, step by painstaking step, as if I were a designated companion, zig-zagging along the bumpy path of their meandering narrative, all the way to the rocky bottom.

Witnesses can exhibit some astonishing lapses of memory, or when asked questions aimed at exposing their exploits, can lash out, grabbing at tangential non sequiturs to help them stay upright, appear more dignified. Transcribers must tail them as they do so, and they must keep up, no matter how loose or rapid or frenzied or nonsensical their talk. We must be faithful to their every stumble, every ‘eh’ and ‘um’ and ‘hmm’, every repeated ‘Can you repeat that question?’, every stubbornness and last reluctance.

You might think this kind of typing sounds straightforward, even mechanical, and yet, paradoxically, it is not. In my experience, transcribing a court proceeding is to fight against a very deep-seated impulse to impose a more rational, reasonable, imaginative, creative and grammatical conditioning onto what people under duress in a witness box tell their interlocutors. Perhaps because I’m a writer, this corrective impulse is doubly ingrained in me, and it has taken an immense effort to completely abandon this improving reflex while I type. Because transcribing is passive, it is anti-writing, utterly devoid of creative effort, and yet it can also be showered with gifts, the same gifts certain great writers clearly have bestowed on them when they work well, which is the unexpected, priceless jewellery of language produced by pressured human thinking. Transcribing is often a freakishly inspired dictation.

Sometimes, when I’ve completed a transcript, I read it back over – the long (and sometimes longwinded) addresses by barristers, the often breathtaking exchanges between counsel and witnesses – and feel, strangely, that I have had a hand in composing it. Many times I look at a finished transcript and groan and think: oh, if only I could write material as weird and beautiful as this! Unadulterated human speech, spoken in a court room, is the stuff of the very best, and worst, dreams. It is aural paradise, and hell. It shares some of the characteristics of the Freudian psychoanalytic process – the witness’s venting (id) versus the barrister’s passive-aggressive humiliations (superego).

Sometimes, I feel I am worsening a person’s evidence by typing it, because if they were were allowed to merely speak, their words could simply evaporate, captured only on a recording that will take some poor soul days and weeks to listen to and recover. Does anyone, other than a transcriber, ever listen to these recordings? My transcript, however, can be word-searched in a matter of seconds by anyone needing, instantly, to pinpoint where a person’s evidence turned tragically, culpably interesting. In converting courtroom speech to text, I am fast-tracking whatever outcome awaits a defendant – jail, suspended sentence, freedom, an irreparably damaged reputation, possibly even justice.

While working, typists communicate together in a sober, professional fashion about the matter at hand, but they are also understandably rebellious and quick-witted to help alleviate the often disturbing content in the work. I’ve been forced to stop typing because I was laughing so much reading a transcriber’s sly comment about anything from the appearance of a wig, to the sometimes challenging names of witnesses. On one case I worked on, a typist logging on to find the case involved ‘hot-tubbing witnesses’ – a term for multiple witnesses giving evidence concurrently, and a major headache for a transcriber – posted, ‘Oh please! Nope. Haven’t drunk enough coffee to be coping with hot-tubbing’, before promptly logging off again. Nurses, lawyers and psychiatrists are known for their black humour; I’d add court transcribers to that list. Since working as a transcriber, my hunger for laughter has really ramped up.


I HAVE READ that the company I work for records 16,000 hours of court recordings per month, which gives you an idea of the number of transcribers required to turn every last spoken exchange in court into transcript documents, the courts’ vital records essential to the functioning of the justice system. I’ve also read how much certain barristers whose words I transcribe are paid: north of $5,000 a day. (That figure is not a transcriber error.) If I type like a demon, I can sometimes make $25 an hour.

Certain matters, heard in higher courts, come to the transcriber with the recorded video of the proceedings, which means I can sit at home and watch the proceeding take place in court too, as if hovering just above the bar table. This makes it far easier to identify who’s who in court, and who is speaking, of course, than just listening in to voices.

When I listen to audio, on its own, the voices of those in court come to me through four channels of audio, which I can manipulate on my computer for volume and speed, and which I can isolate should I need to listen to any of the speakers individually. I’m supplied with a written record of who is who in court by the court recorder so that I know which person is on which channel, and I know what each of them says at certain junctures in the proceedings, because their first spoken words are written in a document I access to check against their allocated microphones.

Making mistakes about who is speaking – confusing defence counsel with a prosecutor, say – or what is said, are, obviously, serious transcribing errors, for which there are repercussions. (Make too many errors, and your rate of pay will drop or, worse, you will be dropped.) It’s something I live in fear of. I check, and check again, that I have matched the barristers’ names with their respective court roles and, for the purposes of accuracy and the carriage of justice, the words they speak. My greatest anxiety, however, is that I will mishear or mistype a witness or defendant’s words. Once, I couldn’t hear whether the defendant – a man with a subdued, hangdog voice – said ‘guilty’ or ‘not guilty’ during his arraignment on charges of having sex with an underage girl. I listened to him on repeat, like a person possessed, until I finally heard him articulate the word ‘guilty’. On one occasion, I asked a member of my family passing my home office to put on my headphones and listen to a word I couldn’t hear, only to discover they could hear it the very first time and with crystalline clarity. Human hearing is fallible, and subjective. Sometimes I hear only what it is I want to hear.

In court, the microphones are allocated like this: the judge, being top dog, is on channel one, the witness on two, the prosecutor on three, defence counsel on four – which is hard not to interpret as a power hierarchy. Knowing who is on what microphone is essential for the transcriber, but normally it only takes me a minute or two to recognise which person is speaking simply by becoming attuned to the characteristics of each voice. Within seconds of listening, I feel I know the players in court, as I might know members of my own family. Each individual has his or her vocal tics, a different rhythm of speech, a tiny idiosyncratic inflection on a certain vowel, an occasional stutter, a telling use of vocabulary, a tone tending to warm or cool, jocular or leaden, friendly or aggressive. All these cues paint a picture in my mind of what sort of man or woman they are, how seriously they take their responsibilities, how inured or fed up they are, whether they are good at fighting, ambitious, persuasive, and if they are from a high- or lowbrow background. Voices, transcribing reinforces to me, are uncanny barometers of personality.

It may seem a reductive piece of stereotyping, but my experience as a transcriber has told me that this, too, is an undeniable truth: judges, for the most part, sound as if they have been raised in homes and schooled in institutions where intonation, cadence, pronunciation and, more importantly, attitude, are still indicators, and products, of a certain class, a very particular demographic. I think I expected the judiciary to sound different, more ‘of the people’, by 2019.

Still, of all the speakers I listen to it’s most often the judges I actually enjoy transcribing. Perhaps it’s because it’s their function to interject and cut through, sometimes in a high-minded way, or with a sparky gruffness, a weary impatience, even rudeness, demanding prosecutors and defence counsel and witnesses be clearer, stay focused, justify an argument or angle of attack – but in almost a year’s worth of work as a transcriber it is the judges I most look forward to listening to. There is one judge, in Queensland, I’d like to write to. I’d tell him that when he retires from the bench he should find a casting agent. What he says, and how he says it, suggests he is the most intelligent, cultured and empathetic man; his manner of speaking is aural gold. I would pay money to see him perform in a theatre.


BECAUSE I HAVE judges’ voices deep in my ears, I believe I can hear – in their expressions, breathing, the movements they make in their chair – the weight of the work they do. I do not envy them. I fancy, as I follow and type what they say, that I can hear their intellects snap and click, hear their emotions brought to heel, their humour breaking through.

Which brings me to the content of what I transcribe. I can’t give much detail about what I’ve transcribed, because I signed a contract forbidding disclosure of it but, in short, here are three things, out of a year’s work, which I doubt I’ll ever expunge from my consciousness. They are from cases now in the public domain, allowing me to mention them, but I think they are likely to be drowned out in the ceaseless tsunami of court evidence heard, never to be spoken of again.

Case one: a male defendant, a trusted neighbour who, it was heard, had sexually abused a young boy throughout the years he babysat him, and who, it was said, had removed the doorknob on the child’s bedroom door to prevent the child’s siblings from coming into the room to see why he spent so much time there. Case two: the sound, deep in my ears, of the tortured, near-hysterical voice of a witness, a relative of a man accused of killing his friend after a banal argument had erupted between the men on a camping trip. Case three: the tremor in the voice of a barrister outlining the charges against a man who had allegedly repeatedly sexually assaulted two of his teenage daughters.

Recently, I read that the judge presiding over the matter of a Palestinian student allegedly raped and murdered in Melbourne had requested the worst details in the transcript be redacted to protect the young woman’s family from the trauma of reading them. I thought his was a merciful and moral decision. Then I wondered about the transcriber who typed that transcript. Most transcribers are women. She would possibly have been at home, working on her own, not even present in court – the invisible, shadow companion of every last harrowing word uttered about those unspeakable acts. There is nothing, not even black humour, that can help a transcriber through that.

Get the latest essay, memoir, reportage, fiction, poetry and more.

Subscribe to Griffith Review or purchase single editions here.

Griffith Review