In the rejection of the application, a great deal is made of the fact that defence accepted the words ‘at the start we made a pact’ as a valid transcription at the time of the trial, and are only opposing it now.
Indeed, with hindsight, it would be easy to argue the defence did a rather poor job in relation to this audio. At the time, however, they did the best they could under the principles established by the law for the handling of indistinct audio.
They objected strenuously to the detective’s transcript. Presumably they would have called a phonetics expert to support their case if they had been aware of this as a possibility.
But that is rare.
Then and now, the normal thing is for defence and prosecution listening to the audio while reading the transcript. This inevitably results in considerable ‘priming’ of both sides (you can find other examples under Problems).
That is one of many reasons we need proper processes for the evaluation of speech evidence.
It is also the reason this fresh evidence makes a difference
The fresh evidence now available in this case is an expert report establishing that the phrase ‘at the start we made a pact’ is an inaccurate transcript.
By its nature, fresh evidence involves information that was not available at the time.
Surely the fact the defence lawyers did not take the fresh evidence into account at the time does not negate the value of the fresh evidence itself? We don’t disallow fresh DNA evidence on the grounds that inaccurate eye witness evidence was accepted at the original trial.
What about the father himself – shouldn’t he have been able to state what he really said?
This is one of those bits of ‘common knowledge’ about speech that seems like a truism but is actually not valid.
Just as we can sometimes fail to decipher our own handwriting when we find a scribbled note we wrote months or years earlier, it is surprisingly common to be unable to make out your own words in a recorded conversation – especially but not only when the audio is of poor quality.
Since few people have the experience of attempting to transcribe recordings of their own conversations, it is easy to maintain the belief that recognising your own words should be easy. It is not.
Here’s my experience
Just recently, I recorded a talk I gave and edited it (for this very website). Although I was editing it just a day or two after giving the talk, there were quite a few places where I was unable to decipher particular phrases I myself had said. And though the quality of the audio in that recording is far from perfect (sorry about that) – I think you will agree it is a heck of a lot better than the covert recording in this case.
Here’s the experience of a police officer
I have in my possession a transcript of a conversation involving an undercover police officer, showing many instances where the officer was unable to make out his own words, and some where he accepted plausible but inaccurate interpretations. Again the audio quality, though poor, was nothing like as bad as the audio in the pact case.
So what about the father in our case study?
The transcription ‘at the start we made a pact’ seemed plausible enough according to the father’s memory of a conversation in which (according to him) he was reminding his son he would stand by him even if he did not turn himself into police.
Remember this is less than 2 seconds in a 38-minute recording, and he is reviewing the police transcript many months after the conversation took place. Also consider that at the time he reviewed the transcript, he did not know the weight that would be given to to this one phrase in the trial.
For these and other reasons, the general principle with speech evidence as with all other evidence should be that it is up to the prosecution to prove their version is accurate, not left for the defence to prove it is not.